# Brigada.dev, full content > IT outsourcing and software development studio in Prishtina, Kosovo. This file contains the full text of our service pages, client case studies, author profiles, and journal index, for AI assistants that prefer one document over crawling. Short version: https://brigada.dev/llms.txt Sitemap: https://brigada.dev/sitemap.xml Case study index: https://brigada.dev/work All content below is written by named people at Brigada. Author and source URL are given for each page. You are welcome to quote and cite this material. ======================================================================== # AI Development Services URL: https://brigada.dev/services/ai-development Author: Endrit Saiti, Co-Founder and CTO (https://brigada.dev/about/endrit-saiti) Last updated: 2026-10-14 Short answer: We build production AI systems from Prishtina, Kosovo: LLM pipelines that turn messy input into structured output, retrieval augmented generation over your own data, agents that act in real systems, and the evaluation harnesses that make prompt changes safe. Four are live, including a CJIS hosted pipeline used by US law enforcement in six states. Most AI projects die somewhere between the demo and the first real user. We build the unglamorous parts that decide whether yours survives. A notebook that answers ten questions correctly is not a product. We learned that the slow way, on systems where being wrong has consequences: a police report that goes to a prosecutor, a revenue decision across a hotel portfolio, a voice assistant people talk to every morning. We came to AI work sideways. We were already running production Laravel and Python for clients with uptime commitments and regulators to answer to. When those clients started asking for AI features, the interesting question was never whether a model could do the task. It was what happens when the model is confidently wrong at two in the morning on a Sunday, and who finds out first, you or your customer. ## What we build We work on four shapes of problem. Most projects are a mix of two or three of them. - LLM pipelines that take messy input (field notes, transcripts, documents, form data) and produce structured output a downstream system can trust. - Retrieval augmented generation over your own content, with retrieval treated as a search problem first and a prompt problem second. - Agents that take actions in real systems: booking, routing, updating records, escalating to a human when confidence drops. - Evaluation harnesses, because the only way to change a prompt safely is to know what broke. ## Why do AI projects stall at eighty percent? Evaluation. Almost nobody builds it up front, and it is the reason AI projects stall at eighty percent good enough. Without an eval suite every prompt change is a guess. Someone tweaks a system prompt to fix one customer complaint, ships it, and quietly regresses four behaviours nobody was watching. Six weeks later the team has stopped touching the prompt at all, because they cannot predict what will happen. We build a graded corpus early, usually a few hundred real cases with known good answers, and wire it into CI. After that, changing the prompt becomes an ordinary engineering task with a pass rate attached. It is dull work and it is the difference between an AI feature you can improve and one you are afraid of. ## How should you choose a model for production? Model choice gets framed as a quality question. In production it is mostly a cost and latency question, and those are product decisions rather than engineering ones. A two stage pipeline where a small fast model handles the easy ninety percent and a larger one handles the rest will often beat a single expensive model on every axis that matters. Cheaper, faster, and easier to debug because you can see which stage failed. We default to that shape and only reach for the big model where it earns its place. ## Can AI run on regulated or private infrastructure? Not every client can send data to a third party API, and for some that is a legal position rather than a preference. We have shipped into a CJIS compliant environment for US law enforcement, onto Swiss hosting with a self hosted text to speech stack for a client who wanted nothing leaving the country, and onto ordinary EU infrastructure for everyone else. If your constraint is regulatory, say so early, because it changes the architecture rather than the deployment script. ## Related work Four of these are live right now. You can go and look at them. ### Police Reports AI (AI / Law Enforcement) A patent pending pipeline that turns an officer's field notes into a complete, regulator compliant report. Officers get back roughly two and a half hours per shift. Running in departments across six US states, hosted in a CJIS compliant environment. Stack: Laravel, React, Python, LLM pipeline, PostgreSQL, AWS Case study: https://brigada.dev/work/police-reports-ai ### Vector Guest (Hospitality / AI) Harold OS, the AI layer running aparthotel portfolios: revenue, demand, guest journey and ground team coordination. Agents plus RAG over a vector database, with LangFuse for tracing. A dedicated Brigada team ships weekly alongside the founding operator. Stack: Laravel, LLM agents, RAG, Vector DB, LangFuse Case study: https://brigada.dev/work/vector-guest ### Prime Tech Group (AI / Consumer Apps) Oxodin, a voice first agentic assistant for a Swiss AI company, live on the App Store and Google Play. Self hosted text to speech, Swiss hosting, and an agent layer that has to feel instant because people speak to it. Stack: React Native, Expo, Laravel, LLM agents, Self-hosted TTS Case study: https://brigada.dev/work/prime-tech-group ### Walter (Sustainability / Web) A US IT asset disposal business where the AI work is quieter: an assisted content pipeline feeding a Statamic CMS, aimed at lead generation rather than headlines. Stack: Laravel 12, Statamic 6, AI content pipeline Case study: https://brigada.dev/work/walter ## Frequently asked questions Q: Do you fine tune models or use APIs? A: Usually APIs, with strong retrieval and evaluation around them. Fine tuning earns its place when you have a large volume of consistent labelled examples and a narrow task. Most clients who ask for fine tuning actually need better retrieval, and we will tell you that before taking the work. Q: How do you stop the model hallucinating? A: You constrain what it can say and you check what it said. In practice that means structured output rather than free text wherever possible, retrieval that grounds answers in your own documents, and a validation layer that rejects or escalates responses failing a rule. Then an eval suite tells you the actual rate rather than your impression of it. Q: Can you run models on our own infrastructure? A: Yes. We have shipped into CJIS compliant hosting for US law enforcement and onto Swiss infrastructure with a self hosted speech stack. Self hosting costs more in engineering time and usually more in compute, so it is worth doing when you have a regulatory or contractual reason rather than a general preference. Q: How long until something is in production? A: Our typical gap between kickoff and first deploy is under two weeks, though that first deploy is deliberately small. A useful AI feature in front of real users is more often six to twelve weeks, and most of that goes into evaluation, edge cases and the fallback path rather than the model call itself. Q: What does an AI project cost to run? A: It depends far more on your traffic shape than on the model price list. We model token cost per request early, because the difference between a naive single model design and a staged one is often five to ten times on the monthly bill at the same output quality. ======================================================================== # Web Development Services URL: https://brigada.dev/services/web-development Author: Endrit Saiti, Co-Founder and CTO (https://brigada.dev/about/endrit-saiti) Last updated: 2026-10-14 Short answer: We build custom web platforms in Laravel, with React or Vue through Inertia, on PostgreSQL or MySQL, hosted on AWS or Laravel Cloud: SaaS products, internal platforms, marketplaces and content sites. Work starts with the data model, the first production deploy usually lands within two weeks, and you own the code from the first commit. Most of what we do is web platforms. Laravel on the back, React or Vue on the front, and a lot of opinions about what happens in between. We build the kind of software a business runs on rather than the kind it markets with. Credentialing and payroll for a healthcare staffing company. A salary exchange engine that has to satisfy HMRC. A nationwide property data platform refreshing lead data daily. Products where being wrong is expensive and being slow is worse. The stack is deliberately boring. Laravel, PostgreSQL or MySQL, Inertia with React or Vue, Tailwind, hosted on AWS or Laravel Cloud. Boring means we can hire for it, hand it over, and debug it at speed two years later. Every hour not spent fighting an exotic framework is an hour spent on the thing you actually asked for. ## What we build Four recurring shapes, and most engagements are one of them. - SaaS products, from the first paying customer through to multi tenant billing and role based access. - Internal platforms that replace a stack of spreadsheets and disconnected tools with one system people will actually use. - Marketplaces and two sided platforms, including payments, payouts and the operational tooling nobody scopes at the start. - Content driven sites where editors need to change everything without opening a ticket. ## We start with the data model The schema is the part you cannot cheaply change later. Interfaces get redesigned constantly and nobody minds. Get the data model wrong and you will be writing migrations against production for the next two years, apologising to your own team. So the first week of a build is usually not visual. It is us asking uncomfortable questions about edge cases you were hoping to defer. What happens when a shift is cancelled after payroll runs. What a refund does to a payout that already left. Whether two records can genuinely be identical. That conversation is cheap now and enormously expensive in month nine. ## Shipping beats staging We deploy early and often, and we would rather put something small in front of real users than perfect something in a staging environment nobody visits. The typical gap between kickoff and first production deploy on our projects is under two weeks. That deploy is usually unimpressive on purpose. It exists so the pipeline, the environment, the monitoring and the rollback path all get exercised while the stakes are low, instead of on launch night. ## Can we take the code back in house later? Some clients keep us for years. Avilog has had us running their engineering for six. Others want to build an internal team and take the code back, and that is a legitimate outcome we plan for rather than resist. In practice that means readable code over clever code, tests where they earn their keep, documented decisions, and no bus factor of one. If handing the project over would be painful for you, that is also a sign the codebase is worse than it should be. ## Related work A sample of web platforms we have shipped and still run. ### Oly Care (Healthcare / SaaS) End to end healthcare staffing: credentialing, shift operations and integrated payroll in one product. Launched in ten weeks and replaced a six tool stack for every pilot agency. We are the engineering team of record on a multi year retainer. Stack: Laravel, Vue 3, Inertia, PostgreSQL, Stripe, AWS Case study: https://brigada.dev/work/oly-care ### Niche Data (Real Estate / Data) A nationwide US property lead platform with daily updated distress data, foreclosures, probates and skiptracing, plus CRM automation. Search runs through Algolia because the data volume made ordinary querying too slow to be useful. Stack: Laravel, Inertia, Vue, PostgreSQL, Algolia Case study: https://brigada.dev/work/niche ### Kinsail (HR / Fintech) A UK workplace nursery scheme built from scratch: salary exchange, an employer dashboard and HMRC compliant reporting. Sign ups convert without support calls, which for a product this regulated was the actual goal. Stack: Laravel, Livewire, Tailwind, MySQL, Stripe, AWS Case study: https://brigada.dev/work/kinsail ### SEOMag Links (Growth / SEO) A premium link building marketplace for agencies, with sponsored placements and an automated order flow. Stripe Connect handles payouts to portal owners, which is the part that quietly makes or breaks a marketplace. Stack: Laravel, Inertia, React, PostgreSQL, Stripe Connect Case study: https://brigada.dev/work/seomag-links ### Evergreen Valleys (Real Estate) A bilingual marketing platform for a sustainable residential development, with 3D rendered house models. Every model and page is editable by the sales team without a developer, which was the entire brief. Stack: Laravel 11, Filament v3, Tailwind, Vite Case study: https://brigada.dev/work/evergreen-valleys ### Walter (Sustainability / Web) The web platform for a US IT asset disposal and e-waste business: CMS, AI assisted content operations and lead generation, on a retainer. Stack: Laravel 12, Statamic 6, Tailwind v4, Forge Case study: https://brigada.dev/work/walter ## Frequently asked questions Q: Why Laravel rather than Node or Rails? A: Because we are genuinely fast in it and the ecosystem covers most of what a business application needs without bolting together five libraries. If you already run Node or Python and want to stay there, we work in those too. Helix Flag is Python and FastAPI. We would rather match your team than win an argument. Q: Do you work with our existing codebase? A: Often, yes. Several engagements started as a rescue or a hardening pass on someone else's code. We will usually ask for a paid discovery week first so we can give you an honest assessment rather than an optimistic one. Q: Can you take over a project that has stalled? A: Yes, and it is more common than greenfield work. The first thing we do is get it deploying reliably, because a project you cannot ship is not a code problem yet, it is a process problem. Q: Who owns the code? A: You do, from the first commit. It sits in your repository under your organisation and we work in it. There is no scenario where ending the relationship means losing access to your own software. ======================================================================== # Mobile App Development Services URL: https://brigada.dev/services/mobile-development Author: Endrit Saiti, Co-Founder and CTO (https://brigada.dev/about/endrit-saiti) Last updated: 2026-10-14 Short answer: We build iOS and Android apps, mostly in React Native with Expo, dropping to native code where a feature needs it, and we usually build the Laravel backend behind them too. Offline use and store review are planned from the start. A focused first release typically takes three to four months, including backend and store submission. Getting an app built is the easy half. Getting it through review, keeping it working without signal, and shipping updates without breaking anyone is the rest of it. We build mobile products where the phone is the primary way people use the thing, not a companion to a website. That changes the engineering. Connectivity is not guaranteed, the update cycle is controlled by two companies who are not you, and a crash rate that would be tolerable on the web will get you uninstalled. Most of our work is React Native with Expo, sharing a codebase across iOS and Android and dropping into native modules where it genuinely matters. We are not religious about it. When a feature needs native performance or a platform API the bridge handles badly, we write native. ## Offline is a design constraint, not a feature If your users are pilots, drivers, field engineers or market traders, they will be offline regularly and they will not accept a spinner as an answer. Building for that means deciding early what the phone is allowed to be the source of truth for, how conflicts resolve when two devices edited the same record, and what the user sees while a sync is pending. Retrofitting offline support into an app that assumed connectivity is close to a rewrite, which is why we ask about it in the first conversation. ## Store review is part of the schedule App Store and Play Store review is not a formality and it is not predictable. We plan for it. Builds go through TestFlight and internal tracks early, so the first submission is not also the first time anyone has seen a release build. Payments, subscriptions, account deletion, privacy labels and permission prompts are the usual rejection reasons. All of them are cheaper to handle correctly at build time than during a rejected release with a launch date already announced. ## The backend is usually the real project A mobile app is a client. Almost every mobile engagement we take on is really a backend and API project with a mobile client attached, and pricing it any other way sets everyone up for disappointment. We build both sides, usually Laravel behind the app, which means the API is designed for the screens that exist rather than reverse engineered from a generic REST layer that fits nothing well. ## Related work Mobile work that is live in the stores or in daily field use. ### Prime Tech Group (AI / Consumer Apps) Oxodin, a voice first agentic personal assistant for a Swiss AI company, live on both the App Store and Google Play. React Native and Expo on the client, Laravel and an agent layer behind it, with self hosted text to speech on Swiss infrastructure. We continue as the engineering team. Stack: React Native, Expo, Laravel, LLM agents, Self-hosted TTS Case study: https://brigada.dev/work/prime-tech-group ### Avilog (Aviation / SaaS) An aviation logbook used by thousands of pilots, six years in production. Offline first by necessity, because pilots log flights where there is no signal, then export regulator ready reports for EASA, FAA and CAA once they are back on the ground. Stack: Laravel, Nuxt, PostgreSQL, Redis, AWS, Offline-first mobile Case study: https://brigada.dev/work/avilog ### Street Wallet (Fintech / Payments) Cashless QR payments for traders, small businesses and corporate programmes across South Africa, pulling Apple Pay, Samsung Pay, SnapScan and Zapper into one wallet with next day settlement. We have been the embedded engineering partner since early 2026. Stack: QR payments, Wallet integrations, Merchant onboarding Case study: https://brigada.dev/work/street-wallet ## Frequently asked questions Q: React Native or fully native? A: React Native for most products, because one codebase across two platforms is a real cost saving and the performance ceiling is higher than its reputation suggests. Fully native when the app is graphics heavy, deeply tied to platform APIs, or when you already have native teams to maintain it. Q: Do you handle App Store and Play Store submission? A: Yes, including the parts nobody enjoys: privacy labels, data safety forms, review responses and staged rollouts. We have shipped to both stores and would rather own that process than hand you a build and wish you luck. Q: Can you take over an existing app? A: Usually. The honest answer depends on how the state management and API layer were built. We do a paid audit first and tell you plainly whether continuing or rewriting is cheaper, including when the answer is not the one that earns us more work. Q: How long does a first version take? A: A focused first release is typically three to four months including backend, store submission and a real beta. Anything promising six weeks is either very small or is not counting review, payments and the offline path. ======================================================================== # DevOps and Cloud Services URL: https://brigada.dev/services/devops-cloud Author: Endrit Saiti, Co-Founder and CTO (https://brigada.dev/about/endrit-saiti) Last updated: 2026-10-14 Short answer: We design and run AWS infrastructure, CI/CD pipelines, observability and on-call for products where downtime matters, including CJIS compliant hosting for US law enforcement and Kubernetes behind a UK healthcare platform. Our bias is the smallest infrastructure that meets your real availability requirement, with Kubernetes only where the workload justifies it. We run infrastructure for products where downtime has a phone number attached to it. That tends to make you conservative in useful ways. A lot of DevOps work sold to small and mid sized companies is architecture theatre: a Kubernetes cluster for an application that would be happier on two servers and a managed database. We have inherited several of those and the pattern is consistent. The cluster is impressive, nobody understands it, and the person who built it has left. Our bias is towards the smallest infrastructure that meets your actual availability requirement, with complexity introduced only when something forces it. Sometimes that is genuinely Kubernetes, as it is for Oly Care. Sometimes it is Forge and a well configured server, and admitting that is worth more to you than a diagram. ## What we take on - AWS architecture and migration, including the unglamorous cost review that usually pays for the engagement. - CI/CD pipelines that run tests, build assets and deploy without a human holding a runbook. - Containerisation with Docker, and Kubernetes only where the workload justifies it. - Observability: logs, metrics, traces and error tracking wired up so an incident starts with evidence rather than guesswork. - Hardening and compliance driven hosting, including CJIS compliant environments and single country requirements. - On-call, with a rota and an escalation path rather than a hopeful Slack message. ## Why do most production outages happen? Most outages we get called into are not caused by exotic failures. They are caused by nobody knowing who is responsible, no alerting on the thing that broke, and a deploy process risky enough that the team avoids shipping the fix. So we start with the boring layer. Alerts that fire on symptoms users feel rather than on CPU graphs. A deploy that can be rolled back by one person under pressure without reading documentation. A written escalation path with names on it. None of this is interesting to talk about and it removes most of the three in the morning phone calls. ## Security work we do by default Databases and caches bound to localhost rather than every network interface. Firewalls actually verified rather than assumed. SSH keys audited and rotated when people leave. Security updates applied on a schedule instead of during an incident. None of that is advanced. It is just the set of things that turns up in every audit we run on infrastructure we did not build, including our own when we go looking honestly. ## Related work Infrastructure we currently run or have hardened. ### Oly Care (Healthcare / SaaS) AWS with Kubernetes behind a healthcare staffing platform handling credentialing, shift operations and payroll. Multi year retainer, which means we live with our own architectural decisions rather than handing them over and leaving. Stack: AWS, Kubernetes, PostgreSQL, Laravel Case study: https://brigada.dev/work/oly-care ### Police Reports AI (AI / Law Enforcement) A CJIS compliant hosting environment on AWS for a platform used by US law enforcement across six states, where the compliance boundary shapes the architecture rather than sitting on top of it. We carry the on-call. Stack: AWS, CJIS-hosted, PostgreSQL, Python Case study: https://brigada.dev/work/police-reports-ai ### Avilog (Aviation / SaaS) Six years of continuous production on AWS with Redis and PostgreSQL, serving thousands of pilots. Long enough that we have migrated it, upgraded it and been paged by it, which is the only honest way to learn what an architecture costs. Stack: AWS, PostgreSQL, Redis, Laravel Case study: https://brigada.dev/work/avilog ### Helix Flag (Fintech / Risk) Platform hardening for a collaborative B2B risk intelligence product on Python and FastAPI, with Sentry for error tracking, as the system moved from working to scaling. Stack: Python, FastAPI, Sentry, Stripe Case study: https://brigada.dev/work/helix-flag ## Frequently asked questions Q: Do we need Kubernetes? A: Probably not. Most products under serious load are still fine on managed services and a small number of well configured servers. Kubernetes earns its complexity when you have many services, real elasticity requirements or a platform team to own it. We will say so when the answer is no. Q: Can you take over infrastructure nobody currently understands? A: Yes, and it is a large share of what we get asked to do. We start with an audit: what exists, what is exposed, what is costing money for nothing, and what would happen if the one person who set it up were unavailable. You get that document whether or not you continue with us. Q: Do you provide on-call? A: For clients where we run the platform, yes, with a rota and a defined escalation path. We do not offer on-call for systems we have not been allowed to fix, because being paged for a problem you are not permitted to solve helps nobody. Q: Can you host inside a specific country or compliance regime? A: We have delivered CJIS compliant hosting for US law enforcement and Swiss only hosting including a self hosted speech stack. Tell us the constraint at the start, because it changes the design rather than the deployment. ======================================================================== # IT Outsourcing in Kosovo URL: https://brigada.dev/it-outsourcing-kosovo Author: Ardit Xhaferi, Co-Founder and CEO (https://brigada.dev/about/ardit-xhaferi) Last updated: 2026-10-14 Short answer: Brigada.dev is a twelve person software team in Prishtina, Kosovo, working Central European Time for clients in twelve countries. We offer dedicated teams, fixed scope project builds and embedded engineers, hold ISO 9001 and ISO/IEC 27001 certification, and price so the same engineers stay on your project. Kickoff to first production deploy is typically under two weeks. We are thirteen people in Prishtina building software for companies in twelve countries. Here is honestly how that works, including the parts harder than the sales pitch suggests. Outsourcing has a reputation problem and most of it is earned. The common story goes like this. You hire an agency, you get a good team for the first month, then the strong engineers rotate onto a newer account and you are left with whoever is available. Communication degrades into status reports. Nobody on the vendor side actually cares whether the product works. We started Brigada in 2024 partly because we had watched that happen from the inside. The way we operate is a reaction to it. Small team, people stay on their projects, and the founders are in the code and in the client calls rather than only on the contract. ## Why outsource to Kosovo? Kosovo is in Central European Time, which puts us in the same working day as all of Europe and gives us a solid overlap with the US East Coast in our afternoons. For UK and EU clients that means an ordinary working relationship with no asynchronous handoff ritual. For US clients it means your morning is our afternoon, and standups actually happen live. The talent pool is younger than Western Europe and it is genuinely good. Kosovo has a large population under thirty, English is widely spoken to a professional standard, and the engineers we work with have been building for international clients from the start rather than adapting to it later. The cost difference is real, and we would rather be plain about it than pretend it is not part of the reason you are reading this page. Rates here are meaningfully below London, Zurich or San Francisco. That is not because the work is worth less. It is because the cost of living in Prishtina is different. The reputational gap is the part Kosovo actually has to close, and the honest answer to it is an audit rather than an adjective. We hold ISO 9001:2015 for quality management and ISO/IEC 27001:2022 for information security, both issued in September 2026 and both covering software development, consulting and the operation of what we build. The certificate numbers and the certificates themselves are on our certifications page. ## How much does IT outsourcing in Kosovo cost? We quote per engineer per month for a dedicated team, and fixed for a project build with a defined scope. We will send you a rate card rather than make you sit through a discovery call to find out whether you can afford us. What actually moves the number is not the technology. It is the seniority mix, whether you need on-call, and how much compliance load the work carries. A team with a named on-call rota and a CJIS or GDPR boundary costs more to run than the same headcount building an internal tool, and any vendor quoting you the same rate for both has not thought about it. The warning we would give about the whole category: the cheapest quote you receive is usually the one that has not understood the problem. Under-quoting is how agencies win work and it is also how engagements go wrong in month four, because the only way back to profitability is to move your strong engineers onto a newer account. We price to be able to keep the same people on your project, which is the thing you are actually buying. ## How does Kosovo compare to other nearshore countries? You are probably comparing us against Poland, Romania, Serbia, North Macedonia or Ukraine, and it is a fair comparison to make. Poland and Romania have deeper senior benches and a longer track record, and they cost more for it. If your requirement is a fifty person delivery organisation with an EU member state on the contract, they are the safer answer and we would tell you so. Ukraine has enormous depth and talent, with a risk profile you already understand. Serbia and North Macedonia are the closest comparison to us on both cost and timezone. Where Kosovo wins is the middle of that range: EU working hours without EU pricing, a young talent pool that has been building for foreign clients from the start, and English at a professional standard across the board. Where it loses is scale and recognition. There are fewer of us, the market is younger, and nobody has heard of it. If a vendor from an unfamiliar country is itself a risk your board will not accept, that is a legitimate reason to go elsewhere and no amount of writing on this page changes it. ## How are outsourcing engagements structured? Three models, and we will tell you which one fits rather than selling you the most expensive one. - Dedicated team. A named group working only on your product, joining your standups and your tooling. This is what Vector Guest and Oly Care have, and it works best for anything long running. - Project build. A defined scope with a start and an end, which suits a first version or a specific platform. Kinsail and Evergreen Valleys were built this way. - Embedded engineers. One or two of our people inside your existing team, reporting into your leads. Street Wallet works like that. ## What we ask from you This is the part most outsourcing pages leave out, and it is the biggest predictor of whether an engagement goes well. We need one person on your side who can make decisions, and we need them reachable. Not a committee, not a weekly sync where questions queue up for six days. Projects that struggle almost always struggle because the answer to a blocking question took four days to arrive, not because the engineering was hard. We also need honesty about constraints early. Regulatory requirements, an existing system nobody wants to touch, a launch date already promised to a board. None of those are problems on day one. All of them are problems in month four. ## What the work usually is Most of what we are outsourced to build is ordinary and load bearing: Laravel and React platforms, mobile apps, the AWS underneath them, and the operational tooling nobody demos but everybody depends on. A growing share of it is AI, and it arrives the same way every time. A client we already run production systems for asks whether a model can do some part of their job. The answer is usually yes and the interesting work is everything after that: retrieval quality, evaluation, cost per run, and what the system does when the model is confidently wrong. We have LLM pipelines, retrieval and agents in production for law enforcement in the US, aparthotel portfolios in the UK and a voice assistant in Switzerland. We mention it here because AI is the one area where outsourcing decisions go wrong for a specific reason: it is the easiest field in software to look competent in. If that is the part you are hiring for, the AI development in Kosovo page goes into what to ask us and every other vendor you are comparing. ## The team you get Thirteen people. Two founders who are still hands on, engineers across Laravel, React, Vue, Python, React Native and AWS, plus DevOps and QA. Everyone works on a small number of projects rather than being spread across eight. You will know their names. They will be in your Slack. When someone goes on holiday you will hear about it in advance from them, rather than discovering it through silence. ## Related work Twelve clients across twelve countries. A few of the longer relationships: ### Avilog (Sweden) Six years in production and we still run the engineering team. The best evidence we can offer that people stay on projects here, because the same relationship has outlasted most agency contracts several times over. Stack: Long-running engineering partner Case study: https://brigada.dev/work/avilog ### Oly Care (United Kingdom) Engineering team of record on a multi year retainer for a UK healthcare staffing platform. Launched the first version in ten weeks and have been the backbone of the product since. Stack: Dedicated team, Multi-year retainer Case study: https://brigada.dev/work/oly-care ### Vector Guest (United Kingdom) A dedicated full time Brigada team shipping weekly with the founding operator, who opened their entire stack to us. That level of access only happens when the working relationship is genuinely close. Stack: Dedicated product team Case study: https://brigada.dev/work/vector-guest ### Police Reports AI (United States) Product, design, engineering and on-call for a platform used by US law enforcement in six states. A US client working with a Kosovo team on software with a compliance boundary, which is the case people assume cannot work. Stack: Product, Engineering, On-call Case study: https://brigada.dev/work/police-reports-ai ## Frequently asked questions Q: What timezone do you work in? A: Central European Time. We overlap the full working day with the UK and EU, and roughly 14:00 to 18:00 CET with the US East Coast. Several of our engineers shift later deliberately for US accounts. Q: Is the whole team in Kosovo? A: Yes, based in Prishtina. That is deliberate. It means the team knows each other, reviews each other's code, and can solve something in a room when it matters. Q: Are you certified to any recognised standard? A: Yes. Brigada Group L.L.C. holds ISO 9001:2015 (certificate MTS-52567) for quality management and ISO/IEC 27001:2022 (certificate MTS-52568) for information security, issued by Invicta Assurance International on 16 September 2026 and valid to 15 September 2027 on a three year cycle with annual surveillance audits. The certified scope is the provision of information technology, software development, consulting and related professional services. Both certificates can be downloaded from brigada.dev/certifications. Q: How do you handle contracts and intellectual property? A: You own the intellectual property and the code from the first commit, held in your own repository. Contracts are straightforward and we are used to working under NDAs and client paper rather than insisting on our own. Q: What happens if we want to stop? A: You keep everything and we help with handover. We would rather end an engagement cleanly and be recommended than lock someone into a relationship they have outgrown. Q: Do you work with startups or only established companies? A: Both. Kinsail and Oly Care were early stage when we started. Avilog and Walter are established businesses. What matters more than stage is whether there is someone who can make decisions quickly. Q: How quickly can you start? A: Our typical gap from kickoff to first production deploy is under two weeks. Actually starting depends on current capacity, and if we cannot staff you properly we will say so instead of taking the work and figuring it out later. Q: How much does IT outsourcing to Kosovo cost? A: Meaningfully less than London, Zurich or San Francisco, because the cost of living in Prishtina is different rather than because the work is worth less. We quote per engineer per month for a dedicated team and fixed for a defined project build, and we will send a rate card rather than require a discovery call first. Seniority mix, on-call and compliance load move the number more than the technology does. Q: How does Kosovo compare to Poland, Romania or Ukraine for outsourcing? A: Poland and Romania have deeper senior benches, a longer track record and EU member state status, and cost more for it. Ukraine has real depth with a risk profile you already understand. Serbia and North Macedonia are our closest comparison on cost and timezone. Kosovo sits in the middle: EU working hours without EU pricing, at the cost of scale and name recognition. Q: Do you do AI development as well? A: Yes, and it is a growing share of the work. LLM pipelines, retrieval and agents in production for US law enforcement, UK hospitality and a Swiss voice assistant. The AI development in Kosovo page covers how we approach it and what to ask any vendor you are comparing us against. ======================================================================== # AI Development in Kosovo URL: https://brigada.dev/ai-development-kosovo Author: Endrit Saiti, Co-Founder and CTO (https://brigada.dev/about/endrit-saiti) Last updated: 2026-10-14 Short answer: AI development in Kosovo is applied engineering rather than model research: LLM pipelines, retrieval and agents built by engineers who already run production systems for foreign clients. Rates sit meaningfully below London or Zurich. Kosovo is outside the EU, so personal data needs standard contractual clauses. Brigada has AI systems live in the US, UK and Switzerland. If you are reading a page about AI development in Kosovo, you are weighing cost against risk. That is the right thing to be weighing, so this page is mostly about the risk. The cost part is easy to establish and everyone selling to you will lead with it. Engineering rates in Prishtina are meaningfully below London, Zurich or the Bay Area, and for AI work specifically that gap matters more than usual, because AI projects tend to need more iterations than anyone budgets for. Cheaper iterations mean you can afford to be wrong a few times, which is the only way this work actually gets done. The risk part is the real question. AI is the easiest field in software to look competent in and one of the hardest to be competent in. A demo takes an afternoon. A system that stays correct when a real user does something nobody anticipated takes months, and almost nothing on a vendor website tells you which one you are buying. What follows is our attempt to give you something you can check. ## Does Kosovo have the talent for AI work? Kosovo does not have a deep bench of people who have trained foundation models. Almost nowhere does, and if that is what your project needs then you want a research lab and a much larger budget, not an agency in the Balkans. What Kosovo does have is a large population of engineers under thirty who came up building production systems for foreign clients, which turns out to be the skill that AI products actually consume. Most of the difficulty in a real AI system is not the model. It is retrieval quality, schema design, queue behaviour under load, evaluation, cost control, and what happens on the failure path. That is ordinary hard software engineering, and it is what the people here are good at. We came to AI work from that direction rather than from research. We were already running Laravel and Python in production for clients with uptime commitments and regulators before any of them asked for an AI feature. When they did, the interesting question was never whether a model could do the task. It was what happens when it is confidently wrong at two in the morning, and who finds out first, you or your customer. ## Is Kosovo in the EU, and does it matter for GDPR? This is the objection that actually kills deals, so we would rather put it near the top than bury it in a FAQ. Kosovo is not an EU member state and does not have a European Commission adequacy decision. In practice that means personal data reaching a team here is a transfer to a third country, handled with standard contractual clauses and a data processing agreement. That is a paperwork exercise your counsel has almost certainly done before, not a blocker, but it is real and you should hear it from us rather than discover it in a security review. Operationally we usually make the question smaller instead of arguing about it. Client data stays in EU regions on AWS, or in one country only where that is contractual. We have built for a Swiss client whose requirement was absolute, running a self hosted speech stack rather than calling an American API, because nothing could leave Switzerland. For AI work specifically the same logic extends to model choice: where a provider boundary is a problem, the pipeline gets designed around it from the start rather than retrofitted after legal reads the architecture diagram. We are not a law firm and will not give you a compliance opinion. What we will do is build the system so that satisfying your compliance people means showing it to them. ## What should you ask an AI development company? If you are comparing us against three other companies in Prishtina, these are the questions that separate them. We would rather you asked everyone than took our word for it. - Show me your evaluation setup. Not the demo, the graded corpus and the pass rate. If there is not one, every prompt change on your project will be a guess. - What runs in production right now, with real users, that you built? Ask how long it has been live. Six weeks and two years are different answers. - What does it cost per month to run, and how did you get it down? A vendor who has never optimised inference cost has never operated anything at volume. - What happens when the model is wrong? A confident answer here, with an escalation path and a human in it, is the strongest signal you will get. - Who owns the code and the prompts? The answer should be you, in your repository, from the first commit. ## How does an AI engagement start? Usually with a two or three week piece of work rather than a signed year. Enough to build the retrieval layer or one pipeline end to end, wire up an eval harness, and put a number on quality that both sides can look at. It is the cheapest way for you to find out whether this works, and for us to find out whether the problem is what everyone assumed it was. From there most clients move to a dedicated team or a defined project build. Our typical gap from kickoff to first production deploy is under two weeks, and if we cannot staff you properly we will say so rather than take the work and figure it out later. ## Related work AI systems we have built from Prishtina, in production with real users: ### Police Reports AI (United States) An LLM pipeline turning field notes into regulator compliant reports for US law enforcement in six states, on CJIS hosted infrastructure. A compliance boundary this strict is the case people assume a Kosovo team cannot work behind. Stack: LLM pipeline, Laravel, Python, CJIS-hosted Case study: https://brigada.dev/work/police-reports-ai ### Vector Guest (United Kingdom) Agents and retrieval running revenue, demand and guest journey across aparthotel portfolios. Vector database, RAG and LangFuse tracing, shipped weekly with a dedicated team. Stack: LLM agents, RAG, Vector DB, LangFuse Case study: https://brigada.dev/work/vector-guest ### Prime Tech Group (Switzerland) Oxodin, a voice first agentic assistant on iOS, Android and web for a Swiss AI company. Self hosted text to speech and Swiss hosting, because no data could leave the country. Stack: LLM agents, Self-hosted TTS, React Native, Swiss hosting Case study: https://brigada.dev/work/prime-tech-group ### Walter (United States) An AI assisted content pipeline behind the web platform of a US IT asset disposal business, where the AI is a working part of lead generation rather than a feature on a roadmap. Stack: AI content pipeline, Laravel 12, Statamic 6 Case study: https://brigada.dev/work/walter ## Frequently asked questions Q: Are there real AI companies in Kosovo? A: Yes, and the number has grown quickly since 2024. Most are small, most do applied work rather than research, and quality varies a lot. The useful filter is not the company size but whether they have an AI system in production with real users and an evaluation suite behind it. Ask to see both. Q: What does AI development cost in Kosovo? A: Meaningfully less than London, Zurich or the Bay Area, and the gap is wider than the quality gap. For AI work that matters more than it does for ordinary development, because these projects need more iterations than anyone budgets for. We quote per engineer per month for a dedicated team, or fixed for a defined build. What moves the number is seniority mix, on-call, and compliance load rather than the technology itself. Q: Is Kosovo in the EU, and does that break GDPR compliance? A: Kosovo is not an EU member and has no adequacy decision, so personal data reaching us is a third country transfer handled under standard contractual clauses and a data processing agreement. It does not break compliance, it adds paperwork your counsel has done before. We keep client data in EU regions, or in a single country where that is the requirement. Q: Which models and providers do you work with? A: Whichever the constraints point at. Claude and GPT class models for most work, smaller or open weight models where cost, latency or a data boundary demands it, and self hosted components where nothing may leave a jurisdiction. We default to a two stage shape where a small fast model handles the easy majority and a larger one handles the rest. Q: Can you work with our existing AI team? A: Yes. Embedding one or two engineers inside an existing team is one of our three engagement models. It tends to work well when your team knows the domain and needs production and evaluation capability rather than more model expertise. Q: Who owns the models, prompts and code you build? A: You do, from the first commit, in your own repository. That includes prompts, evaluation corpora and infrastructure code. If you stop working with us you keep all of it and we help with handover. Q: What timezone does your AI team work in? A: Central European Time, from Prishtina. A full working day of overlap with the UK and EU, and roughly 14:00 to 18:00 CET with the US East Coast. Several engineers shift later deliberately for US accounts. ======================================================================== # DevOps Outsourcing in Kosovo URL: https://brigada.dev/devops-outsourcing-kosovo Author: Ardit Xhaferi, Co-Founder and CEO (https://brigada.dev/about/ardit-xhaferi) Last updated: 2026-10-14 Short answer: DevOps outsourcing to Kosovo gets you AWS, Kubernetes, CI/CD and a named on-call rota run from Prishtina in Central European Time: full overlap with UK and EU working hours and afternoon overlap with the US East Coast. Brigada operates CJIS hosted infrastructure for US law enforcement and is certified to ISO/IEC 27001:2022 for information security. Outsourcing DevOps is not like outsourcing a feature. If a feature is late you ship next week. If your infrastructure is down at three in the morning, the only question that matters is who picks up. That is why this page spends most of its length on on-call rather than on a tools list. Everyone offering DevOps from anywhere runs AWS, Kubernetes, Terraform and CI/CD, ourselves included, and listing them tells you nothing. What separates vendors is what happens during the hours you are asleep and what happens the morning after. We run production for a US law enforcement platform on CJIS hosted infrastructure, a UK healthcare product on Kubernetes, and a Swedish aviation platform that has been live for six years. All of it operated from Prishtina. The parts worth explaining are the ones people get wrong. ## Can a team in Kosovo cover on-call across timezones? Central European Time is the whole argument for DevOps outsourcing to Kosovo, and it cuts both ways. For a UK or EU client our working day is your working day, so an incident at eleven in the morning gets a live human immediately rather than a ticket. For a US client our afternoon covers your morning, and our night covers your night, which is the part that needs saying out loud. Nobody should pretend a team in one timezone provides follow the sun coverage on its own. What we run instead is a named rota with a defined escalation path and paging that works at any hour, staffed by people who deliberately shift later for US accounts. That is a real 24/7 commitment rather than a marketing one, and the difference is whether the person who wakes up has the access and the context to actually fix it. The question to ask any DevOps vendor, us included: who specifically is on call tonight, what is their escalation path, and what was your last incident. A vendor who cannot answer all three in one breath is selling you a tools list. ## What we run from Prishtina Four shapes of work. Most engagements are two or three of them. - AWS infrastructure design, migrations and the cost review that usually pays for the work. Terraform for anything that should be reproducible. - Kubernetes where it earns its place, and deliberately not where it does not. Plenty of products are better served by boring managed services and we will say so. - CI/CD pipelines so test, build and deploy happen without a human holding a runbook, which is the single highest leverage thing most teams are missing. - Observability and on-call: logs, metrics, traces and error tracking, so an incident starts with evidence rather than with someone guessing. ## Is DevOps outsourcing to Kosovo compliant with GDPR? Most infrastructure outsourcing conversations that stall do not stall on capability. They stall when someone asks where the data lives and who can reach it. Kosovo is not an EU member and has no adequacy decision, so personal data reaching a team here is a third country transfer handled under standard contractual clauses and a data processing agreement. That is paperwork, not a blocker, and it is a conversation we have had many times. Beyond the paperwork the practical answer is to design the boundary in from the start. EU regions on AWS for most clients. Single country hosting where it is contractual, which we have done for a Swiss client for whom nothing could leave Switzerland. CJIS hosted infrastructure for a US law enforcement platform, which is a stricter bar than most European clients will ever need and is the clearest evidence we can offer that the ceiling here is higher than the sector's reputation. Underneath all of that sits a certified information security management system. We hold ISO/IEC 27001:2022, certificate MTS-52568, issued by Invicta Assurance International in September 2026, alongside ISO 9001:2015 for quality management. It covers access control, client data handling, supplier risk and incident response across the engineering team, which is usually what a security questionnaire is really asking about. The certificates are downloadable at brigada.dev/certifications. ## How much does DevOps outsourcing to Kosovo cost? DevOps engineers are the scarcest and therefore most expensive part of the Kosovo talent pool, so expect to pay above the general software engineering rate here. It will still be meaningfully below London or Zurich. The more useful number is usually the other one. Most AWS accounts we are handed the keys to are spending between twenty and forty percent more than they need to, almost always on the same handful of things: oversized instances chosen during a launch and never revisited, storage nobody owns, environments running at full size overnight, and inter-region traffic nobody has ever looked at. A cost review is a week of work and it frequently pays for a quarter of the engagement. We quote per engineer per month for ongoing operations and fixed for a defined migration or audit. On-call is priced separately and explicitly, because a vendor who folds it into a day rate without mentioning it has not planned to actually staff it. ## Related work Production infrastructure we run from Prishtina: ### Police Reports AI (United States) CJIS hosted AWS infrastructure for a platform used by US law enforcement in six states. The strictest compliance boundary we operate behind, run by a team in Kosovo. Stack: AWS, CJIS-hosted, PostgreSQL, On-call Case study: https://brigada.dev/work/police-reports-ai ### Oly Care (United Kingdom) Kubernetes on AWS behind a UK healthcare staffing platform handling credentialing, shifts and payroll. Multi year retainer with a dedicated team and a named on-call rota. Stack: AWS, Kubernetes, Laravel, PostgreSQL Case study: https://brigada.dev/work/oly-care ### Avilog (Sweden) Six years of continuous production for a Swedish aviation platform serving thousands of pilots. The most honest uptime evidence we have is that the same team has run it that long. Stack: AWS, PostgreSQL, Redis, Six-year engagement Case study: https://brigada.dev/work/avilog ### Helix Flag (Fintech) Python and FastAPI risk intelligence platform with Sentry wired in from the start, so incidents begin with a stack trace rather than a customer email. Stack: Python, FastAPI, Sentry Case study: https://brigada.dev/work/helix-flag ## Frequently asked questions Q: Can a DevOps team in Kosovo really provide 24/7 on-call? A: With a named rota, defined escalation and paging that works at any hour, yes. What a single team in one timezone cannot honestly claim is follow the sun coverage with everyone working normal hours. We staff a real rota and several engineers shift later deliberately for US accounts. Ask any vendor who is on call tonight and what their last incident was. Q: How much does DevOps outsourcing to Kosovo cost? A: Above the general Kosovo software engineering rate, because DevOps engineers are the scarcest part of the talent pool here, and still meaningfully below London or Zurich. We quote per engineer per month for ongoing operations and fixed for a defined migration or audit. On-call is priced separately and explicitly. Q: What timezone do you cover? A: Central European Time from Prishtina. A full working day of overlap with the UK and EU, and roughly 14:00 to 18:00 CET with the US East Coast, plus an on-call rota outside those hours. Q: Which cloud platforms do you work with? A: AWS primarily, with Terraform for anything that should be reproducible and Kubernetes where it earns its place. We have also run single country and self hosted setups where a data boundary made a public cloud region unacceptable. Q: Can you take over infrastructure someone else built? A: Yes, and it is most of what we are asked to do. It normally starts with an audit covering access, backups, restore testing, cost and the parts nobody has documented. The restore test is the one that finds problems, because plenty of teams have backups nobody has ever tried to restore. Q: Is our data safe with a team outside the EU? A: Kosovo has no EU adequacy decision, so it is a third country transfer under standard contractual clauses with a data processing agreement. Operationally we keep data in EU regions, or one country only where that is contractual. Our information security management system is certified to ISO/IEC 27001:2022 (certificate MTS-52568). We also run CJIS hosted infrastructure for a US law enforcement client, which is a stricter bar than most European engagements require. Q: How quickly can you take over an on-call rota? A: Not immediately, and be wary of anyone who says otherwise. We need access, a runbook walkthrough and at least one incident shadowed before we are the people you page. Two to three weeks is typical to do that properly. ======================================================================== # The Kosovo Tech Sector, in Numbers URL: https://brigada.dev/kosovo-tech-talent Author: Ardit Xhaferi, Co-Founder and CEO (https://brigada.dev/about/ardit-xhaferi) Last updated: 2026-10-14 Short answer: Kosovo's software sector is small and export led. STIKK counts roughly 120 software businesses employing about 3,000 IT professionals; 85 percent of companies export and 39 percent work only for foreign clients. An engineer costs roughly 1,400 euro a month to employ. The population is Europe's youngest, but senior depth is thin and brain drain is real. Every agency page about Kosovo makes the same three claims: young population, good English, low cost. All three are broadly true and none of them are numbers. This page is the numbers, including the ones that are not flattering. We are a Kosovo company, so treat this as interested rather than neutral. What we have tried to do is source every figure and say when it is from, because the statistics that circulate about this sector are frequently years old, occasionally mutually contradictory, and sometimes just wrong. One salary aggregator we checked while writing this page lists the average Kosovo senior software engineer at 2,642 euro per year, which is not a real number. Where sources disagree we have said so rather than picked the flattering one. Where a figure is old we have dated it. If you are building a business case for outsourcing to Kosovo, you want the version with the holes visible. ## How many software developers are there in Kosovo? Smaller than the marketing suggests. STIKK, the Kosovo ICT association, puts the software side of the industry at roughly 120 IT software businesses employing around 3,000 IT professionals. Estimates for the wider ICT and BPO sector run to around 2,800 registered businesses, but that broader figure sweeps in telecoms, hardware retail and call centres, so it is not the number you want if you are looking for software engineers. Three thousand software professionals is the figure to plan against. For scale, that is a rounding error next to Poland or Ukraine. It means Kosovo can staff your team, and it means Kosovo cannot staff a fifty person delivery organisation for you next quarter. Anyone telling you otherwise is describing a country they have not counted. The sector is overwhelmingly export facing. STIKK's IT Barometer finds 85 percent of companies export services or products, and 39 percent work exclusively for international markets. That matters more than the headcount, because it means the working norms here are already foreign client norms rather than something a vendor has to retrofit for you. ## Which countries does Kosovo export software to? Germany, the United States and Switzerland are the three largest markets for Kosovo technology firms. The United Kingdom sits fourth, having grown from roughly 4 million to 13.9 million euro in recent years. Exports of telecommunications, computer and information services were about 99 million euro in 2021, which was 5.1 percent of Kosovo services exports. That figure is the most recent clean one we could source and it is now several years old, so treat it as a floor rather than a current number. The practical read for a buyer: if you are in Germany, the US, Switzerland or the UK, a Kosovo team has almost certainly worked with your market, your contract conventions and your invoicing expectations before. If you are somewhere else, ask specifically. ## How experienced are Kosovo developers? Kosovo has the youngest population in Europe. Median age is around 32, which is the lowest on the continent, and depending on the source somewhere between 55 percent under 30 and 70 percent under 35. Sources genuinely disagree here, and the trend is worth noting too: average age rose from about 29.9 in 2011 to 34.8 in 2024, driven by falling birth rates and emigration. That demographic produces a specific shape of talent pool, and it is important to understand the shape rather than just the size. There are a lot of capable engineers with three to eight years of experience who have been building for foreign clients since they started. There are considerably fewer with twenty years behind them, because twenty years ago this industry did not exist here. So Kosovo is strong on capable mid-level and senior engineers who work well in an international team, and thin on the grey-haired principal architect who has personally survived four database migrations. If your project needs the latter, budget for the search or bring that person yourself. ## How much do software engineers cost in Kosovo? Domestic salary benchmarks put a Kosovo software engineer at roughly 1,300 euro per month gross, with employer contributions adding about 100 euro, for a total employment cost near 1,400 euro per month as of 2026 guidance. Broader IT roles are often quoted in a 700 to 1,200 euro range depending on seniority. Do not convert those into an agency rate and assume you have been overcharged. Domestic salary is one input. An agency rate also carries employer contributions, recruitment, equipment, holiday and sick cover, management, QA, DevOps, and the margin that funds keeping the same engineers on your project instead of rotating them onto a newer account. A quote that looks close to the raw salary figure is a quote from someone who has not priced retention, and you will meet the consequences of that in month four. The honest framing is that Kosovo gives you meaningfully lower cost than London, Zurich or the Bay Area, at broadly similar cost to Serbia or North Macedonia, and below Poland and Romania. It does not give you offshore Asia pricing, and a Kosovo vendor quoting offshore Asia pricing is telling you something about who they will staff you with. ## What the numbers do not show Four things worth knowing before you build a business case on any of the above. - Brain drain is real and it is the sector's structural problem. Engineers leave for Germany and Switzerland, and the ones who stay have options. Retention is a genuine differentiator here rather than a slogan, and you should ask any vendor how long their engineers have actually been with them. - Kosovo is not an EU member and has no adequacy decision. Personal data reaching a team here is a third country transfer requiring standard contractual clauses. It is paperwork rather than a blocker, but it is paperwork your legal team will find. - Kosovo does not appear in the EF English Proficiency Index at all, so anyone citing an English ranking for Kosovo is citing something that does not exist. English here is genuinely strong in the tech sector, but the honest evidence for that is talking to the engineers, not an index. - The ecosystem is young in absolute terms. StartupBlink ranks Kosovo around 98th globally and Prishtina around 824th among cities. Tech Park Prishtina, inaugurated recently, is targeting 300 qualified jobs and 60 companies over five years, which tells you both that there is real institutional momentum and how small the base still is. ## Related work We are one of the roughly 120 software companies in those figures. What that looks like in practice: ### Avilog (Sweden) Six years in production with the same engineering team. In a sector where brain drain is the structural problem, a six year client relationship is the most useful retention statistic we can offer. Stack: Six-year engagement, Laravel, Nuxt, AWS Case study: https://brigada.dev/work/avilog ### Police Reports AI (United States) A US law enforcement platform on CJIS hosted infrastructure, built from Prishtina. Evidence that the compliance ceiling for a Kosovo team is higher than the sector's reputation suggests. Stack: LLM pipeline, CJIS-hosted, Laravel, Python Case study: https://brigada.dev/work/police-reports-ai ### Prime Tech Group (Switzerland) Switzerland is one of the three largest markets for Kosovo tech exports. This is what the high end of that trade looks like: Swiss hosting, self hosted speech, nothing leaving the country. Stack: LLM agents, Swiss hosting, React Native Case study: https://brigada.dev/work/prime-tech-group ### Oly Care (United Kingdom) The UK is the fourth largest and fastest growing market for Kosovo technology firms. A healthcare staffing platform launched in ten weeks and on a multi year retainer since. Stack: Laravel, Vue 3, AWS, Kubernetes Case study: https://brigada.dev/work/oly-care ## Sources - STIKK, Kosovo IT Barometer: https://stikk.org/en/stikk-publishes-kosovo-it-barometer-2021-2022-report/ (company counts, employment, and the 85 percent / 39 percent export figures.) - German Economic Team, Kosovo ICT and BPO sector: https://www.german-economic-team.com/en/newsletter/kosovos-ict-and-bpo-sector-developments-and-outlook/ (services export value for 2021 and sector composition.) - bne IntelliNews, Kosovo ICT industry: https://www.intellinews.com/booming-ict-industry-gives-kosovo-hope-for-the-future-298400/ (export market ranking and UK growth from roughly 4m to 13.9m euro.) - Our World in Data, Kosovo population and demography: https://ourworldindata.org/profile/population-demography/kosovo (age structure and the 2011 to 2024 median age trend.) - Playroll, cost of hiring employees in Kosovo: https://www.playroll.com/employment-cost/kosovo (2026 salary and employer contribution guidance.) - StartupBlink, Kosovo startup ecosystem: https://www.startupblink.com/top-startups/kosovo (global and city ecosystem rankings.) - STIKK, Tech Park Prishtina: https://stikk.org/en/inaugurohet-tech-park-prishtina-moment-historik-per-te-ardhmen-e-teknologjise-ne-kosove/ (five year targets for jobs and hosted companies.) ## Frequently asked questions Q: How many software developers are there in Kosovo? A: STIKK, the Kosovo ICT association, estimates roughly 120 IT software businesses employing around 3,000 IT professionals. Wider figures of around 2,800 ICT businesses include telecoms, hardware and call centres, so 3,000 software professionals is the number to plan against. Q: How much does a software developer cost in Kosovo? A: Domestic salary benchmarks put a software engineer near 1,300 euro per month gross with around 100 euro of employer contributions, for roughly 1,400 euro total monthly employment cost as of 2026 guidance. Agency rates sit above that because they also carry recruitment, cover, management, QA, DevOps and retention. Q: Is Kosovo good for outsourcing software development? A: For a team of a few engineers to about twenty, in Central European Time, at below Western European cost, yes. For a fifty person delivery organisation staffed next quarter, no, because the whole country has around 3,000 software professionals. The honest answer depends entirely on the size and seniority you need. Q: What is the biggest weakness of the Kosovo tech sector? A: Scale and seniority depth, both caused by how young the industry is, and brain drain to Germany and Switzerland on top of that. There are many capable engineers with three to eight years of experience and comparatively few with twenty. Q: Which countries do Kosovo tech companies work with most? A: Germany, the United States and Switzerland are the three largest markets, with the United Kingdom fourth and growing fastest, up from roughly 4 million to 13.9 million euro in recent years. Q: Do Kosovo developers speak English? A: In the tech sector, generally yes and to a professional working standard, since 85 percent of companies here export and 39 percent work exclusively for international clients. Kosovo is not ranked in the EF English Proficiency Index, so treat any specific English ranking you see quoted for Kosovo as fabricated, and judge it by talking to the engineers instead. ======================================================================== # Hire Developers in Kosovo URL: https://brigada.dev/hire-developers-kosovo Author: Ardit Xhaferi, Co-Founder and CEO (https://brigada.dev/about/ardit-xhaferi) Last updated: 2026-10-14 Short answer: There are three ways to hire developers in Kosovo: employ them directly through a local entity, use an employer of record, or work with an agency. Direct hiring costs near 1,300 euro a month gross plus about 100 euro of contributions, but takes two to three months. Agencies and employers of record take about two weeks. There are three ways to get Kosovo engineers onto your product, and the one most vendors recommend is the one that pays them most. Here are all three, including the two that do not involve us. Most pages ranking for this search are marketplaces, and a marketplace has one answer to every question, which is to use the marketplace. We are an agency, so we have our own bias and you should assume it. The way to read this page is that we have tried to describe the options accurately enough that you can rule us out quickly if we are wrong for you. The three routes are hiring directly, hiring through an employer of record, and working with an agency. They differ on cost, speed, risk and how much management you personally absorb, and the right answer depends mostly on how many people you need and for how long. ## What are the ways to hire developers in Kosovo? Roughly in order of cheapest and slowest to most expensive and fastest. - Hire directly. Cheapest per head and by a wide margin, since you pay salary plus contributions rather than a rate. It requires a Kosovo legal entity or a local employment arrangement, plus you doing the recruiting, the technical screening, the onboarding and the retention in a market you do not know. Right answer if you are hiring five or more people and intend to be here for years. - Employer of record. You choose the person, a third party employs them for a fee, typically a few hundred euro per person per month on top of salary. No entity needed and no long commitment. Right answer when you have already found the person, or you know exactly who you want and just need them payrolled legally. - Agency, which is us. Highest rate per head, and you are not paying for a body. You are paying for a team that already exists and already works together, plus recruitment, cover for holiday and illness, QA, DevOps, code review and the management layer. Right answer when you need capability in weeks rather than quarters, or a mix of skills you cannot justify hiring individually. ## When should you not hire an agency? If you can hire two strong senior engineers in your own city and keep them for three years, do that instead. It is the best outcome available and no agency should tell you otherwise. If you need one specific person for one specific long term role and you have already found them, use an employer of record and skip the margin. If you are building a fifty person engineering organisation, build your own entity, because at that headcount the agency arithmetic stops working in your favour and the whole Kosovo software sector is only around 3,000 people anyway. Agencies earn their rate in a narrower band than agency websites suggest: a few engineers to about twenty, needed soon, on work with a defined arc, where you would rather not carry permanent headcount. Outside that band you are usually better served by one of the other two routes. ## How much does it cost to hire developers in Kosovo? Domestic salary benchmarks put a Kosovo software engineer near 1,300 euro per month gross, plus roughly 100 euro of employer contributions. That is the number people find first and then use to conclude that agency rates are outrageous, so it is worth spelling out what the gap covers. An agency rate carries the salary, the employer contributions, recruitment and technical screening, equipment, holiday and sick cover so your project does not stop when one person is away, QA, DevOps, code review, a management layer, and margin. That margin is what funds keeping the same engineers on your project rather than rotating your strong people onto a newer account, which is the single most common way outsourcing relationships fail. We quote per engineer per month for a dedicated team and fixed for a defined project build, and we will send a rate card rather than require a discovery call first. Seniority mix, on-call and compliance load move the number far more than the technology does. ## How long does it take to hire developers in Kosovo? From signed agreement to an engineer actively shipping, roughly two weeks is the realistic figure for Kosovo across all three routes, and it matches what the staffing platforms quote. Our own typical gap from kickoff to first production deploy is under two weeks. Direct hiring is the exception and takes considerably longer, because you are running a recruitment process in an unfamiliar market before any of that clock starts. Budget two to three months to hire well, and longer if you need senior people, since that is exactly the part of the Kosovo talent pool that is thinnest. If we cannot staff you properly we will say so rather than take the work and figure it out later. That happens a few times a year and it is a better outcome for both sides than the alternative. ## What should you check before signing a vendor? Whether you use us, a marketplace or a competitor down the road in Prishtina, these are the questions that separate them. - How long have your engineers been with you? Brain drain is the structural problem in this sector. Average tenure is the single most revealing number a Kosovo vendor can give you, and the reluctant ones are telling you something. - Who exactly will work on this, and can I meet them before signing? Named people, not a capability deck. The classic failure is a strong team for month one and whoever is available by month three. - Who owns the code and the IP? The answer should be you, in your repository, from the first commit. - What happens if we want to stop? A clean exit with handover should be a normal part of the contract, not a negotiation. - Is the data transfer handled? Kosovo is not in the EU and has no adequacy decision, so a GDPR relevant engagement needs standard contractual clauses and a data processing agreement. A vendor who has not thought about this has not worked with EU clients seriously. ## Related work What hiring us has looked like for other people: ### Avilog (Sweden) Six years and still the engineering team of record. If you only ask a Kosovo vendor one question, ask how long their clients and their engineers stay. This is our answer. Stack: Six-year engagement, Dedicated team Case study: https://brigada.dev/work/avilog ### Oly Care (United Kingdom) First version of a UK healthcare staffing platform live in ten weeks, replacing a stack of six separate tools, then a multi year retainer with a dedicated team. Stack: Dedicated team, Ten weeks to launch Case study: https://brigada.dev/work/oly-care ### Street Wallet (South Africa) Embedded engineers rather than a separate team, working inside the client's existing setup on QR payments and merchant onboarding. The lowest commitment way to work with us. Stack: Embedded engineers, QR payments Case study: https://brigada.dev/work/street-wallet ### Kinsail (United Kingdom) A defined project build with a start and an end: UK workplace nursery scheme with salary exchange and an HMRC compliant reporting engine, built from scratch. Stack: Project build, Laravel, Stripe Case study: https://brigada.dev/work/kinsail ## Frequently asked questions Q: How much does it cost to hire a developer in Kosovo? A: Direct employment runs near 1,300 euro per month gross plus around 100 euro of employer contributions as of 2026 guidance. An employer of record adds a few hundred euro per person per month. An agency rate is higher again because it also covers recruitment, holiday and sick cover, QA, DevOps, management and retention. Q: How long does it take to hire a developer in Kosovo? A: Through an agency or employer of record, roughly two weeks from signed agreement to someone actively shipping. Hiring directly takes two to three months because you run the recruitment yourself in an unfamiliar market, and longer for senior roles. Q: Do I need a legal entity in Kosovo to hire here? A: Only for direct employment. An employer of record or an agency arrangement needs no entity, which is the main reason most companies start with one of those and open an entity later if the headcount justifies it. Q: Should I use an agency or hire directly? A: Hire directly if you need five or more people and intend to be in Kosovo for years, since the per head cost is far lower. Use an agency if you need a few engineers soon, a mix of skills you cannot justify hiring individually, or capacity without permanent headcount. If you have already found one specific person, an employer of record is cheaper than both. Q: Can Kosovo developers work US hours? A: Partly. Kosovo is in Central European Time, which gives a full working day of overlap with the UK and EU and roughly 14:00 to 18:00 CET with the US East Coast. Several of our engineers shift later deliberately for US accounts. US West Coast overlap is genuinely limited and we would rather say that than sell around it. Q: What happens if the engineer is not a good fit? A: You tell us and we replace them, and that is on us rather than billed to you. It is worth confirming this in writing with any vendor, because the answer varies more than you would expect. ======================================================================== # Nearshore Software Development for Europe URL: https://brigada.dev/nearshore-software-development-europe Author: Ardit Xhaferi, Co-Founder and CEO (https://brigada.dev/about/ardit-xhaferi) Last updated: 2026-10-14 Short answer: Nearshore software development means a team close enough in time to share your working day. Brigada works from Prishtina in Central European Time, which gives UK and EU clients a full day of overlap, live standups and same hour code review. Client data stays in EU regions on AWS, or in a single country where that is contractual. Nearshore only means anything if the overlap is real. Ours is the entire working day, from Prishtina, in Central European Time. Offshore development to a timezone eight hours away produces a specific kind of failure. Every question costs a day. Your team writes long clarifying documents instead of asking. Work arrives that answers the question you asked rather than the one you meant, and by the time you see it there is another day of work stacked on top. Nearshore fixes that by removing the delay, not by improving the documentation. We are in Central European Time. If a Berlin, Amsterdam, Stockholm or London team has a question at eleven in the morning, they have an answer at eleven in the morning. ## What does timezone overlap actually buy you? Live standups instead of written status. Pair debugging when something is genuinely hard. A code review that comes back within the hour rather than tomorrow. Decisions made in a fifteen minute call instead of a thread that runs for three days. It sounds minor written down. Over a six month engagement it is the difference between a team that feels like part of your company and a supplier you send requirements to. ## How do you handle GDPR and hosting? For EU and UK clients the data questions come up early and they should. We work to GDPR expectations by default: data minimisation in the schema rather than in a policy document, access controls that match roles, retention and deletion built in rather than promised. On hosting we go where the requirement points. EU regions on AWS for most clients. Single country hosting where that is contractual, which we have done for a Swiss client who wanted nothing leaving Switzerland, including running a self hosted speech stack rather than calling an American API. We are not a law firm and will not pretend to give you a compliance opinion. What we do is build so that satisfying your compliance people is a matter of showing them the system rather than apologising for it. ## Is nearshore better than hiring locally? Honestly, if you can hire two strong senior engineers in your own city and keep them for three years, do that. It is the best outcome and no agency should tell you otherwise. Nearshore makes sense when hiring is slow, when you need a capable team in weeks rather than quarters, when the work has a defined arc, or when you want capacity without permanent headcount. It also makes sense when you need a mix of skills, a backend specialist plus a mobile developer plus someone who knows AWS properly, that you cannot justify hiring individually. ## Related work European clients we work with from Prishtina. ### Kinsail (United Kingdom) A UK workplace nursery scheme with salary exchange and HMRC compliant reporting. Regulated UK payroll logic built end to end by a team in Kosovo, on UK working hours. Stack: Laravel, Livewire, MySQL, Stripe Case study: https://brigada.dev/work/kinsail ### Oly Care (United Kingdom) UK healthcare staffing platform handling credentialing, shifts and payroll. Launched in ten weeks, now on a multi year retainer with a dedicated team. Stack: Laravel, Vue 3, AWS, Kubernetes Case study: https://brigada.dev/work/oly-care ### Prime Tech Group (Switzerland) A Swiss AI company where the hosting requirement was absolute. Swiss infrastructure, self hosted text to speech, nothing leaving the country. Live on both app stores. Stack: React Native, Laravel, Swiss hosting Case study: https://brigada.dev/work/prime-tech-group ### Avilog (Sweden) A Swedish aviation platform we have run for six years, serving thousands of pilots with EASA, FAA and CAA compliant exports. Stack: Laravel, Nuxt, PostgreSQL, AWS Case study: https://brigada.dev/work/avilog ### SEOMag Links (Poland) A marketplace for SEO agencies with sponsored placements, automated order flow and Stripe Connect payouts to portal owners. Stack: Laravel, React, Stripe Connect Case study: https://brigada.dev/work/seomag-links ## Frequently asked questions Q: Is Kosovo in the EU? A: No. Kosovo is not an EU member state, though it is in Central European Time and geographically in the Balkans. For most clients this is a contractual detail rather than an operational one, and we host client data in EU regions or wherever your requirement specifies. Q: How do you handle GDPR? A: By building for it rather than documenting around it: minimal personal data in the schema, role based access, retention and deletion paths that exist in code. We sign data processing agreements and work under client contracts. For a formal compliance opinion you want your own counsel, not us. Q: Can you host everything inside the EU? A: Yes, and inside a single country where that is the requirement. We have done Swiss only hosting including self hosted speech synthesis for a client who could not use US APIs. Q: What languages does the team speak? A: English throughout, to a professional working standard. All code, documentation and communication is in English by default. Albanian is our first language and several of the team speak German. Q: How is this different from offshore development? A: Timezone, mostly, and what that does to how a team behaves. With a full working day of overlap you get live conversation and same hour code review. With an eight hour gap you get documents and delay, and quality suffers in ways that are hard to attribute at the time. ======================================================================== # Nearshore Software Development in the Balkans URL: https://brigada.dev/nearshore-software-development-balkans Author: Ardit Xhaferi, Co-Founder and CEO (https://brigada.dev/about/ardit-xhaferi) Last updated: 2026-10-14 Short answer: Six Western Balkans countries compete for nearshore software work: Serbia, North Macedonia, Bosnia and Herzegovina, Albania, Kosovo and Montenegro. All work Central European Time and none is in the EU. Serbia is by far the largest, with around 115,000 people in ICT. Kosovo, at around 3,000 software professionals, suits teams of a few engineers to about twenty. If you have decided to look at the Balkans for software development, the next question is which country. Here is how the six compare, with the source for every number, written by a company in one of them. We are based in Prishtina, so read this as interested rather than neutral. We have tried to make it useful anyway, because most comparisons of the region are written by a vendor in one country quietly arguing for it, and they rarely show where their numbers come from. Ours are linked at the bottom of the page, with the year each one describes. One warning before any figures. The countries measure their tech sectors differently. Some count software companies, some count the whole ICT sector including telecoms, and Albania's figure includes call centres. The numbers below tell you the order of magnitude and the ranking with some confidence. They do not support precise comparisons, and we have said so wherever it matters. ## What do the Balkan countries have in common? More than the differences suggest. All six work Central European Time, the same as Berlin, Paris and Stockholm and an hour ahead of London, so any of them gives a UK or EU client a full shared working day and gives the US East Coast an afternoon of overlap. None of them is an EU member state, and none has an EU adequacy decision for data protection, so personal data sent to a team anywhere in the region is a third country transfer handled under standard contractual clauses. They also share a shape: tech sectors that are younger and smaller than their Central European equivalents, and that grew up selling abroad. For the Europe wide argument for working in your own timezone at all, see our nearshore software development in Europe page. This page assumes you are past that and choosing between countries. ## How big is the tech workforce in each Balkan country? Serbia is in a different league from the rest. The Statistical Office of Serbia, as reported by the US International Trade Administration, puts the ICT sector at around 115,000 people in more than 4,000 companies, and Serbia's ICT services exports reached a record 4.55 billion euro in 2025. If you need a large delivery organisation inside the region, Serbia is the only country that can plausibly staff one. The middle of the range is North Macedonia and Bosnia and Herzegovina. MASIT, the Macedonian ICT chamber, counts 23,948 people employed in the ICT sector in 2024, with 664 million euro of ICT services exports. The Foreign Trade Chamber of Bosnia and Herzegovina counts more than 1,100 IT companies and more than 12,000 employees, with IT services exports above 800 million convertible marks in 2025. Albania's investment agency reports 27,000 people in ICT and BPO combined, which includes call centres and outsourced back office work, so the software share is considerably smaller. Kosovo and Montenegro are the smallest. STIKK, the Kosovo ICT association, counts roughly 120 software businesses employing around 3,000 IT professionals. For Montenegro we could find a count of around 800 ICT companies but no reliable headcount, and we would rather leave the gap visible than fill it. ## Which Balkan countries are closest to joining the EU? EU status changes nothing about data protection today, since none of the six is a member. It does say something about the direction of each country's regulation and contract law, which matters to some procurement teams. As the European Commission and the European Parliament describe it in 2026: - Montenegro is furthest along: a candidate since December 2010, negotiating since 2012, with every screened chapter opened. - Serbia has been a candidate since March 2012 and has opened 22 of 35 negotiating chapters, two of them provisionally closed. - Albania has been a candidate since June 2014 and has opened all of its negotiating clusters. - North Macedonia applied in 2004 and has been a candidate since December 2005, but no negotiating chapters have been opened yet. - Bosnia and Herzegovina became a candidate in December 2022, and the decision to open negotiations was taken in March 2024. - Kosovo is a potential candidate. It submitted its membership application in 2022 and has had a Stabilisation and Association Agreement with the EU in force since April 2016. ## Where is English strongest in the Balkans? The EF English Proficiency Index is the only cross country measure most buyers will see, so it is worth knowing exactly what it says. In the 2025 edition, which covers 123 countries and regions, North Macedonia ranks 17th, Bosnia and Herzegovina 21st and Serbia 25th, all in the high proficiency band. Albania ranks 42nd, in the moderate band. Kosovo and Montenegro are not ranked at all. That is an absence of data rather than a low score, and anyone quoting a national English ranking for either country is quoting something that does not exist. In our experience English across the Kosovo tech sector is at a professional working standard, but the honest evidence for that is a call with the engineers, not an index. The same advice holds everywhere in the region: the index measures adults who took a test, not the specific people who will be on your project. ## Kosovo vs Albania vs Serbia: which should you choose? It depends mostly on how many people you need, and only then on everything else. For a team of fifty, or a supplier with a long track record and several hundred engineers on staff, Serbia is the realistic answer inside the Balkans, and it has the deepest bench by a wide margin. For ten to twenty engineers, North Macedonia and Bosnia and Herzegovina have more depth than Kosovo or Montenegro and strong English on paper. Albania shares a language with Kosovo and has a larger workforce, though its headline figure is less software specific. Kosovo is the right answer for a narrower but common case: a few engineers to about twenty, from a sector where 85 percent of companies export and 39 percent work only for foreign clients, so the working habits are already built around someone like you. The trade is scale and recognition. There are fewer people, the senior bench is thinner because the industry is young, and the country name is unfamiliar to most boards. If that last point alone is a risk your board will not accept, that is a legitimate reason to choose elsewhere. Whichever country you pick, the vendor matters more than the flag. Ask how long their engineers have been with them, who exactly will work on your product, and who owns the code. Those answers vary far more between companies in the same city than between the countries on this page. ## What the numbers on this page do not tell you Four limits worth knowing before you build a business case on any of the above. - The workforce figures use different definitions and years. Serbia and North Macedonia count the ICT sector, Bosnia and Herzegovina counts IT companies, Albania includes BPO and Kosovo counts software firms. Use them for order of magnitude, not for ratios. - We found no single salary source that measures all six countries the same way, so we have not printed a cost comparison that would look precise and not be. In our experience costs are broadly similar across the region and well below Western Europe, but compare actual quotes rather than country averages. - Export totals show where the sector sells, not how good any individual vendor is. - Population sets the ceiling. By World Bank estimates for 2025, Serbia has about 6.5 million people, Bosnia and Herzegovina 3.1 million, Albania 2.3 million, North Macedonia 1.8 million, Kosovo 1.6 million and Montenegro 0.6 million. A small country can staff your team. It cannot staff a hundred. ## Related work What a Balkan nearshore team looks like in practice, from Prishtina: ### Oly Care (United Kingdom) A UK healthcare staffing platform launched in ten weeks, then a multi year retainer with a dedicated team working UK hours from Kosovo. Stack: Dedicated team, Laravel, Vue 3, AWS Case study: https://brigada.dev/work/oly-care ### Kinsail (United Kingdom) Regulated UK payroll logic, including HMRC compliant reporting, built end to end by a team in Kosovo on UK working hours. Stack: Project build, Laravel, Livewire Case study: https://brigada.dev/work/kinsail ### Prime Tech Group (Switzerland) A Swiss AI company with an absolute hosting requirement: Swiss infrastructure and self hosted speech, delivered from outside the EU without any data leaving Switzerland. Stack: Swiss hosting, React Native, Laravel Case study: https://brigada.dev/work/prime-tech-group ### Avilog (Sweden) Six years in production with the same engineering team, the most useful answer we have to the retention question every Balkan buyer should ask. Stack: Six-year engagement, Laravel, Nuxt Case study: https://brigada.dev/work/avilog ## Sources - European Parliament, The Western Balkans fact sheet (April 2026): https://www.europarl.europa.eu/factsheets/en/sheet/168/the-western-balkans (accession progress across the region, Kosovo's 2022 application and North Macedonia's unopened chapters.) - European Commission, Montenegro: https://enlargement.ec.europa.eu/montenegro_en (candidate status since 2010 and all screened chapters opened.) - European Commission, Serbia: https://enlargement.ec.europa.eu/serbia_en (candidate status since March 2012, 22 of 35 chapters opened and two provisionally closed.) - European Commission, Albania: https://enlargement.ec.europa.eu/albania_en (candidate status since June 2014 and all negotiating clusters opened.) - European Commission, North Macedonia: https://enlargement.ec.europa.eu/north-macedonia_en (application in 2004 and candidate status since 2005.) - European Commission, Bosnia and Herzegovina: https://enlargement.ec.europa.eu/bosnia-and-herzegovina_en (candidate status since December 2022 and the March 2024 decision to open negotiations.) - European Commission, Kosovo: https://enlargement.ec.europa.eu/kosovo_en (potential candidate status and the Stabilisation and Association Agreement in force since April 2016.) - European Commission, adequacy decisions: https://commission.europa.eu/law/law-topic/data-protection/international-dimension-data-protection/adequacy-decisions_en (the list of countries with an adequacy decision, which includes none of the six.) - US International Trade Administration, Serbia ICT market: https://www.trade.gov/country-commercial-guides/serbia-information-and-communications-technology-market (around 115,000 ICT employees in more than 4,000 companies, citing the Statistical Office of Serbia.) - Tanjug, Serbian ICT services exports in 2025: https://www.tanjug.rs/english/economy/231708/jovanovic-serbian-ict-services-exports-exceeded-45-bln-euros-in-2025/vest (ICT services exports of 4.552 billion euro in 2025.) - MASIT, ICT industry growth in 2024: https://masit.org.mk/en/publications-en/the-growth-trend-of-the-ict-industry-continues-in-2024/ (23,948 ICT employees and 664 million euro of ICT services exports in North Macedonia in 2024.) - BIT Alliance, Bosnia and Herzegovina at GITEX Europe 2026: https://www.bit-alliance.ba/en/bosnia-and-herzegovina-presented-at-gitex-europe-2026-in-berlin (more than 1,100 IT companies, more than 12,000 employees and IT exports above 800 million KM, citing the Foreign Trade Chamber.) - AIDA, Albania ICT and BPO fact sheet 2025: https://aida.gov.al/wp-content/uploads/2025/04/AIDA_FactSheet_ICTBPO2025.pdf (27,000 people employed across ICT and BPO in Albania.) - STIKK, Kosovo IT Barometer: https://stikk.org/en/stikk-publishes-kosovo-it-barometer-2021-2022-report/ (around 120 software businesses, 3,000 IT professionals, and the 85 percent / 39 percent export figures.) - US International Trade Administration, Montenegro ICT sector: https://www.trade.gov/country-commercial-guides/montenegro-information-and-communication-technology-ict-sector (around 800 ICT companies in Montenegro.) - EF English Proficiency Index 2025: https://www.ef.com/wwen/epi/ (ranks and proficiency bands, and the absence of Kosovo and Montenegro.) - World Bank, population total: https://data.worldbank.org/indicator/SP.POP.TOTL (2025 population estimates for all six countries.) ## Frequently asked questions Q: Which Balkan country is best for software outsourcing? A: It depends on team size. Serbia has by far the largest ICT workforce, around 115,000 people, and suits large teams. North Macedonia and Bosnia and Herzegovina sit in the middle. Kosovo, with around 3,000 software professionals, suits teams of a few engineers to about twenty. All six work Central European Time. Q: Are any Western Balkan countries in the EU? A: No. Montenegro, Serbia, Albania, North Macedonia and Bosnia and Herzegovina are EU candidates, and Kosovo is a potential candidate that applied in 2022. None has an EU adequacy decision, so personal data sent to a team in any of them is a third country transfer under standard contractual clauses. Q: Kosovo vs Serbia for software development? A: Serbia has the scale: around 115,000 ICT workers against roughly 3,000 software professionals in Kosovo, and ICT services exports of 4.55 billion euro in 2025. Kosovo suits smaller teams that want a sector built around foreign clients, with 85 percent of its tech companies exporting. Both work Central European Time. Q: Kosovo vs Albania for software development? A: They share a language and a working culture. Albania reports 27,000 people in ICT and BPO combined, a figure that includes call centres, and ranks 42nd in the EF English Proficiency Index. Kosovo counts around 3,000 software professionals and is not ranked in the index. Compare the vendors rather than the countries. Q: Which Balkan country has the best English? A: On the EF English Proficiency Index 2025, North Macedonia ranks 17th, Bosnia and Herzegovina 21st and Serbia 25th, all rated high proficiency, with Albania 42nd at moderate. Kosovo and Montenegro are not ranked, which means no data rather than a low score. Q: Is the Balkans cheaper than Poland or Romania? A: Generally yes, and well below Western Europe, but we could not find one salary source measuring all six Balkan countries the same way, so treat any precise country by country cost table with suspicion and compare real quotes. Seniority mix and on-call move the number more than the country does. ======================================================================== # Laravel Developers in Kosovo URL: https://brigada.dev/laravel-developers-kosovo Author: Endrit Saiti, Co-Founder and CTO (https://brigada.dev/about/endrit-saiti) Last updated: 2026-10-14 Short answer: Brigada.dev is a Laravel development company in Prishtina, Kosovo. Most of our engineers work in Laravel daily, on platforms such as Oly Care, Kinsail, Avilog, Niche Data and Vector Guest. We build with Livewire or Inertia, host on Forge, Laravel Cloud or AWS, keep apps on supported versions, and work Central European Time. Laravel is the stack most of our work runs on. This page is what we would want to know if we were hiring a Laravel team in Kosovo, including us. Search for Laravel developers in Kosovo and you mostly get directories and freelance marketplaces. That is fine if you want one person for a few weeks. It tells you very little if you want a team that will still be looking after your application in three years, which is the question that actually decides whether a Laravel project goes well. We are a software studio in Prishtina, and Laravel is the core skill for most of our engineers. It sits behind a UK healthcare staffing platform, a UK workplace nursery scheme with HMRC compliant reporting, a Swedish aviation logbook that has been in production for six years, a nationwide US property data platform, and the AI operating system behind an aparthotel portfolio. The site you are reading is Laravel 12 with Inertia, React and server side rendering. What follows is how we think about the framework, and what we would check before trusting any Laravel team with your product. ## Why hire Laravel developers in Kosovo? Three reasons, and one of them is not cost alone. Kosovo is in Central European Time, so a Laravel team here shares the whole working day with the UK and EU and overlaps the US East Coast in the afternoon. The engineers came up building for foreign clients, so code review, pull requests and written English are the default rather than an adjustment. And Laravel is what we hire for: the seat we opened in September 2026 was for a Laravel developer who has already shipped it to production. The cost difference is real and we will not pretend otherwise. Rates in Prishtina are meaningfully below London, Zurich or San Francisco. We do not publish ours, but we will send a rate card rather than make you sit through a discovery call to find out. The hire developers in Kosovo page sets out what an engineer costs to employ here and what an agency rate carries on top. The limit is scale. The whole Kosovo software sector is around 3,000 people. A team of three to fifteen Laravel engineers is very realistic. Fifty by next quarter is not, from us or from anyone else here. ## What should you look for in a Laravel team? Most Laravel agencies look identical on their websites. These are the questions that separate them, and we would rather you asked everyone than took our word for it. - What Laravel applications have you built that are still in production, and for how long? A team that has lived with its own code for years has learned things a team that hands over at launch never does. Avilog has been in production with us for six years. - Which version is each of those applications on today? An agency whose client apps sit on unsupported releases is telling you how your app will be treated in year two. - Show me a test suite. Not a coverage badge, the tests themselves, and what happens in CI when one fails. - How do you handle queues, scheduled jobs and failed jobs? Most production incidents in a Laravel app start in the background, not on the page someone is looking at. - Who owns deployment and who gets paged? If the answer is the client, the agency is selling you half a service. - Who owns the code? It should be you, in your repository, from the first commit. ## Livewire or Inertia? We run both in production and neither is the right answer everywhere. Kinsail is Laravel and Livewire. Oly Care and Niche Data are Inertia with Vue. SEOMag Links and this website are Inertia with React. Evergreen Valleys uses Filament for everything the sales team edits, and Walter runs on Statamic. Livewire keeps the whole application in PHP and Blade, which means one language, one mental model and fewer moving parts. It suits products that are mostly forms, dashboards and administration, and teams that would rather not run a JavaScript build as a second application. Kinsail is exactly that shape: a regulated product where the interesting logic is on the server and the interface needs to be clear rather than clever. Inertia earns its place when the interface is genuinely rich, when you already have React or Vue skills, or when you want the front end to feel like a single page application without building and versioning a separate API. The cost is that you now maintain two ecosystems and their upgrades. Filament and Statamic are a different decision again: they are how you give non developers control of content and records without building an admin panel from scratch. What we try to avoid is the choice being made by whichever the first developer happened to know. It is a decision about who maintains the product in year three. ## Forge, Vapor or Laravel Cloud? Hosting a Laravel application is a smaller decision than it looks and a harder one to reverse than people expect. Ours currently spread across most of the options. Walter runs on servers managed through Forge. Vector Guest is deployed on Laravel Cloud. Oly Care runs on AWS with Kubernetes, and Police Reports AI sits in a CJIS compliant AWS environment where the compliance boundary decides the architecture. Forge on a well configured server is the boring default and we mean that as praise. You get a server you can reason about, predictable cost and a short path from problem to fix. Laravel Cloud removes most of the server work in exchange for less control, and suits teams who would rather not own infrastructure at all. Vapor runs Laravel serverless on AWS Lambda, which suits sharply spiky traffic, but changes how you think about queues, file storage and anything that runs for a long time. None of the projects on this page run on Vapor, and we would only recommend it for a traffic pattern that needs it. Kubernetes is the option we argue clients out of more often than into. Oly Care needs it. Most Laravel applications are happier on two servers and a managed database, and the DevOps outsourcing in Kosovo page goes into why. ## How do you keep a Laravel application up to date? Laravel ships a major release roughly once a year. Each one gets bug fixes for 18 months and security fixes for two years. In practice that means Laravel 11 stopped receiving security fixes on 12 March 2026, and Laravel 12 receives them until 24 February 2027. An application nobody has upgraded for two years is running on code nobody is patching. We treat upgrades as maintenance rather than a project. One major version at a time, close to its release, with the test suite run before and after, and PHP moved forward in the same rhythm. Laravel 12 was deliberately a light release, and most applications moved to it without changing application code. That is the reward for staying current. The bill arrives when a team skips three versions and has to cross all of them at once, usually under pressure from a security finding. Avilog is the longest example we have. We have migrated it, upgraded it and been paged by it over six years, which is the only honest way to learn what keeping a Laravel application healthy actually costs. ## What does testing look like on a Laravel project? Tests where they earn their keep, not tests for the sake of a percentage. In a Laravel application that mostly means feature tests that hit real routes and assert on what comes back, because they survive refactoring and catch the failures a user would actually see. Unit tests go where the logic is genuinely intricate: a payroll calculation, a pricing rule, a regulatory export. The tests that matter most are the ones that encode a rule nobody should break quietly. On this website, a test fails the build if the XML sitemap and the registered routes ever disagree, because a hardcoded list and a derived one will drift apart the moment nothing checks them. On Kinsail, the reporting engine was built HMRC compliant from the start, because in that category a reporting bug is a compliance incident rather than a ticket. The QA and testing page covers how review, QA and release gates fit together across our projects. ## Can you take over a Laravel project that has stalled? Yes, and it is more common than greenfield work. Several of our engagements started as a rescue or a hardening pass on someone else's code. We usually ask for a paid discovery week first, so we can give you an honest assessment rather than an optimistic one, and you keep the written findings whether or not you continue with us. The first job is almost never a feature. It is getting the application deploying reliably, because a project you cannot ship is a process problem before it is a code problem. After that the usual list is unglamorous: an outdated framework version, a schema that has drifted from the code, logic duplicated in several places, background jobs failing silently, and no tests around the parts that handle money. We work through it in that order of risk, and you see the list before we start. Sometimes the honest answer is that a rewrite is cheaper than a rescue. We will say so when it is, including when the answer is not the one that earns us more work. ## Related work Laravel applications we have built and still run, from Prishtina: ### Oly Care (United Kingdom) Laravel, Vue 3 and Inertia behind a UK healthcare staffing platform covering credentialing, shifts and payroll. Launched in ten weeks, replaced a six tool stack, and we have been the engineering team of record since. Stack: Laravel, Inertia, Vue 3, PostgreSQL, Kubernetes Case study: https://brigada.dev/work/oly-care ### Kinsail (United Kingdom) Laravel and Livewire for a UK workplace nursery scheme: salary exchange, an employer dashboard and HMRC compliant reporting, built end to end. Sign ups convert without support calls. Stack: Laravel, Livewire, Tailwind, MySQL, Stripe Case study: https://brigada.dev/work/kinsail ### Avilog (Sweden) Laravel and Nuxt for an aviation logbook used by thousands of pilots, six years in production with the same team, with EASA, FAA and CAA exports. Stack: Laravel, Nuxt, PostgreSQL, Redis, AWS Case study: https://brigada.dev/work/avilog ### Niche Data (United States) A nationwide property lead platform on Laravel, Inertia and Vue, with daily updated distress data. Search runs through Algolia because the data volume made ordinary querying too slow to be useful. Stack: Laravel, Inertia, Vue, PostgreSQL, Algolia Case study: https://brigada.dev/work/niche ### Vector Guest (United Kingdom) Harold OS, the AI layer running aparthotel portfolios: Laravel underneath tool calling agents and retrieval, deployed on Laravel Cloud, shipped weekly by a dedicated team. Stack: Laravel, Laravel Cloud, LLM agents, RAG Case study: https://brigada.dev/work/vector-guest ### Walter (United States) Laravel 12 and Statamic 6 on Forge for a US IT asset disposal business, with an AI assisted content pipeline, run on a maintenance retainer. Stack: Laravel 12, Statamic 6, Tailwind v4, Forge Case study: https://brigada.dev/work/walter ## Sources - Laravel release notes and support policy: https://laravel.com/docs/12.x/releases (annual major releases, 18 months of bug fixes and two years of security fixes, and the end dates for Laravel 11 and 12.) ## Frequently asked questions Q: Are there good Laravel developers in Kosovo? A: Yes. Most of our own engineers work in Laravel, and engineers here have typically built for foreign clients from the start of their careers, so production habits come with the framework. The sector is small, around 3,000 software professionals, so the realistic team size is a few engineers to about fifteen rather than fifty. Q: How much does it cost to hire Laravel developers in Kosovo? A: Meaningfully less than London, Zurich or San Francisco. We do not publish rates, but we quote per engineer per month for a dedicated team and fixed for a defined build, and we will send a rate card without a discovery call first. Seniority mix, on-call and compliance load move the number more than the framework does. Q: Do you use Livewire or Inertia? A: Both, in production. Livewire for products that are mostly forms and dashboards and should stay in PHP, as on Kinsail. Inertia with Vue or React when the interface is rich or the team already has those skills, as on Oly Care, Niche Data and SEOMag Links. We choose based on who maintains it in year three. Q: Can you upgrade an old Laravel application? A: Yes. We move one major version at a time with the test suite run before and after each step, and bring PHP forward alongside it. Laravel gives each release 18 months of bug fixes and two years of security fixes, so an application left alone for two years is no longer being patched. Q: Where do you host Laravel applications? A: Wherever the requirement points. Forge on a managed server for Walter, Laravel Cloud for Vector Guest, AWS with Kubernetes for Oly Care, and a CJIS compliant AWS environment for Police Reports AI. For most applications a well configured server and a managed database is the right answer. Q: Do Laravel developers in Kosovo work UK or US hours? A: Kosovo is in Central European Time, so UK and EU clients get the full working day. US East Coast overlap is roughly 14:00 to 18:00 CET, and several of our engineers shift later deliberately for US accounts. US West Coast overlap is limited and we would rather say that plainly. Q: Can you work inside our existing Laravel team? A: Yes. Embedding one or two of our engineers with your developers, reporting to your leads and working in your repository, is one of our three engagement models alongside a dedicated team and a defined project build. ======================================================================== # QA and Software Testing Outsourcing in Kosovo URL: https://brigada.dev/qa-testing-outsourcing-kosovo Author: Endrit Saiti, Co-Founder and CTO (https://brigada.dev/about/endrit-saiti) Last updated: 2026-10-14 Short answer: Brigada.dev provides QA and software testing from Prishtina, Kosovo, as part of engineering rather than a separate bench of testers. Work passes code review and QA before it ships, behind CI gates, regression suites and staged releases. We also harden existing codebases, and our quality management system is certified to ISO 9001:2015. We do not sell testing by the hour. We build software where a bug has a cost attached, and testing is how we keep shipping it. If that is what you need from a QA partner in Kosovo, read on. If you need a room of manual testers, we will tell you that too. Software testing outsourcing usually means one of two things. Either a testing company runs scripted test cycles against software someone else wrote, or an engineering team owns quality as part of building and running the product. Both are legitimate. They solve different problems, and buying the wrong one is an expensive way to find out. We are the second kind. QA sits inside our team alongside engineering and DevOps, and every project goes through QA review before release. Most of what we test is our own work: a payroll engine that answers to HMRC, logbook exports that answer to three aviation regulators, a reporting pipeline used by US law enforcement. Some of it is someone else's code that we were brought in to make safe to ship. ## Where does QA sit in the delivery process? Before release, not after it. A change is specified, built, reviewed by another engineer, checked in QA, and only then released. When QA arrives at the end of a project as a separate phase, it finds problems at the moment they are most expensive to fix and least likely to be fixed properly. The other half is what happens when something gets through anyway. Our ISO 9001 audit asked exactly that: how work gets specified, reviewed and corrected, and what we do when a delivery goes wrong rather than what we intend to do. The honest finding was that most of it we already did, and the gap was being able to show we had done it every time. Certification covers the management system rather than any single project, but it does mean the evidence has to keep existing between audits. ## Manual or automated testing? Both, for different jobs. Automated tests carry the regressions: the behaviour that has to stay exactly the same release after release, which a human checking by hand will eventually miss on a busy Friday. Manual testing carries the new: a feature nobody has used yet, an edge case the specification did not imagine, a flow that is technically correct and still confusing. The most useful automated tests we write encode a rule rather than a line of code. On Helix Flag that meant golden record regression tests: a fixed set of risk records with known correct verdicts, which every release has to reproduce exactly. On a regulated product, the equivalent is the calculation or export that a regulator will read. Coverage percentages matter much less than whether those particular tests exist. Mobile adds its own layer. Builds go to TestFlight and internal tracks early, so the first store submission is not also the first time anyone has seen a release build on a real device. ## What is a release gate, and why does it matter? A release gate is a check that has to pass before code can ship, enforced by the pipeline rather than by someone remembering. On Helix Flag we put in CI hard gates so unreviewed code cannot ship, end to end browser tests, and Sentry on both the front and back end so a failure in production starts with a stack trace rather than a customer email. Gates only work if releasing is cheap. That is why our first production deploy on a new project is usually unimpressive on purpose: it exists so the pipeline, the monitoring and the rollback path are exercised while the stakes are low. A deploy that one person can roll back under pressure without reading documentation is part of quality, not a separate DevOps concern. ## How do you test AI features? With an evaluation suite, because an AI feature cannot be tested by checking that a function returns the right value once. We build a graded corpus early, usually a few hundred real cases with known good answers, and wire it into CI. After that, a prompt or model change comes with a pass rate attached instead of a feeling. Without it, every prompt change is a guess. Someone fixes one customer complaint, ships it, and quietly regresses four behaviours nobody was watching. The AI development page goes into this in more depth. ## Can you test and harden a codebase someone else built? Yes. Helix Flag came to us with real customers and fragile foundations: no CI gate, schema drift between the code and the database, a signup and billing flow duplicated in five places, and a browser extension that could disagree with the web app about the same risk check. We audited it before touching anything, then worked through the list. Today releases are gated and regression tested, and the team ships features instead of firefighting. That is the shape we would suggest for any inherited product. An audit first, so you see the risks in order before paying for fixes. Then the gates and the regression tests around whatever handles money, compliance or decisions, because that is where a quiet bug does the most damage. ## When should you hire a dedicated testing company instead? If you need a large pool of manual testers running scripted cycles against software built elsewhere, a broad device lab, or independent testing that has to be separate from the people who wrote the code, a specialist testing company is the better fit. We are not set up to sell testing hours on their own, and we would rather say so than stretch. Where we fit is when you want quality owned by the people building and running the product: a dedicated team or project build with QA inside it, or a hardening engagement on a codebase that has outgrown its safety net. ## Related work Products where testing decides whether the thing can be trusted: ### Helix Flag (United States) Audit first, then CI hard gates, end to end browser tests, Sentry on both ends and golden record regression tests for the risk engine, across a network of more than 3 million risk records. Stack: Regression tests, CI gates, Sentry, Python Case study: https://brigada.dev/work/helix-flag ### Kinsail (United Kingdom) Salary exchange and HMRC compliant reporting, built compliant from the start, because in this category a reporting bug is a compliance incident rather than a ticket. Stack: Laravel, Livewire, HMRC reporting Case study: https://brigada.dev/work/kinsail ### Avilog (Sweden) Exports in the formats EASA, FAA and CAA expect, from one source of truth, for thousands of pilots over six years in production. Stack: Laravel, Nuxt, Offline first, Regulator exports Case study: https://brigada.dev/work/avilog ### Police Reports AI (United States) An LLM pipeline whose output goes to a prosecutor, running in departments across six US states on CJIS hosted infrastructure, where being wrong has consequences. Stack: LLM pipeline, Laravel, Python, CJIS-hosted Case study: https://brigada.dev/work/police-reports-ai ## Frequently asked questions Q: Do you offer software testing outsourcing in Kosovo? A: Yes, as part of engineering rather than as standalone testing hours. QA sits inside our delivery teams, every project goes through QA review before release, and we take on hardening work for existing codebases. If you need a large bench of manual testers on software built elsewhere, a specialist testing company is the better fit. Q: Do you do manual or automated testing? A: Both. Automated regression tests carry the behaviour that must not change between releases, such as calculations, exports and verdicts. Manual testing covers new features and the edge cases a specification did not anticipate. Mobile builds go through TestFlight and internal tracks before store submission. Q: Are you ISO certified for quality? A: Yes. Brigada Group L.L.C. holds ISO 9001:2015 for quality management (certificate MTS-52567) and ISO/IEC 27001:2022 for information security (certificate MTS-52568), issued by Invicta Assurance International in September 2026. Both certificates are downloadable at brigada.dev/certifications. Q: Can you add tests to a codebase that has none? A: Yes, and we start with an audit rather than with coverage. The first tests go around whatever handles money, compliance or decisions, alongside a CI gate so nothing unreviewed ships. Coverage of everything else follows the risk, not a percentage target. Q: How do you test AI and LLM features? A: With a graded corpus of real cases and known good answers, wired into CI, so a prompt or model change comes with a pass rate. It is the only reliable way to change an AI feature without quietly breaking behaviour nobody was watching. Q: What timezone does your QA work in? A: Central European Time from Prishtina, with a full working day of overlap with the UK and EU and roughly 14:00 to 18:00 CET with the US East Coast. ======================================================================== # Certifications URL: https://brigada.dev/certifications Certificates are issued to Brigada Group L.L.C., Str. Xhafer Shala, No. 41, 10000 Prishtina, Republic of Kosovo, by Invicta Assurance International (SHPK). Accredited by the International Accreditation Service (IAS), management systems certification body MSCB-136. IAF MLA member. Certified scope: Provision of Information Technology, Software Development, Consulting and Related Professional Services ## ISO 9001:2015 — Quality Management System Certificate number: MTS-52567 Issued: 2026-09-16 · Expires: 2027-09-15 Reference: IAF code 33-35 Certificate PDF: https://brigada.dev/certificates/brigada-iso-9001-2015.pdf Audited quality management: how work is specified, reviewed, delivered and corrected, and how we handle it when something goes wrong. ## ISO/IEC 27001:2022 — Information Security Management System Certificate number: MTS-52568 Issued: 2026-09-16 · Expires: 2027-09-15 Reference: Technical area B.01-C.01 · Statement of Applicability UBS 23.02.2026 REV00 Certificate PDF: https://brigada.dev/certificates/brigada-iso-iec-27001-2022.pdf Audited information security: access control, client data handling, supplier and infrastructure risk, and incident response across the engineering team. Both certificates run on a three year certification cycle and remain valid provided the organisation continues to comply with the certification requirements and passes the surveillance audit held at least once a year. ======================================================================== # Police Reports AI — case study URL: https://brigada.dev/work/police-reports-ai Industry: AI / Law Enforcement Product: https://www.policereports.ai/ Stack: Laravel, React, Python, LLM pipeline, PostgreSQL, AWS, CJIS-hosted AI-powered report generation platform used by US law enforcement agencies. Patent-pending workflow that turns field notes into complete, regulator-compliant reports in minutes. Outcome: Officers reclaim ~2.5 hours per shift. Rolled out to departments across six states. Report quality up, rework down. ======================================================================== # Oly Care — case study URL: https://brigada.dev/work/oly-care Industry: Healthcare / SaaS Product: https://oly.care/ Stack: Laravel, Vue 3, Inertia, PostgreSQL, Stripe, AWS, Kubernetes End-to-end healthcare staffing platform — credentialing, shift ops, and integrated payroll in one product. Replaced a stack of six separate tools. Outcome: Launched in 10 weeks. Replaced the six-tool stack for every pilot agency. Now the engineering backbone of a growing SaaS business. ## The problem: six tools and data that was stale on arrival Healthcare staffing agencies were running on a stack of six products: an applicant tracking system, a credentialing tracker, a scheduler, a payroll provider, an invoicer, and a compliance dashboard. Each one held its own copy of the same nurses, shifts and clients. Six copies of the truth means none of them is current. By the time a credential update or a shift change had made its way from one tool to the next, the data was stale before it reached the admin who had to act on it. ## One data model, one product team We took on the whole surface area rather than a slice of it. One data model underneath candidates, credentials, shifts, payroll and compliance, so a change in one place is a change everywhere. One product team owning all of it, so there is nobody to hand the problem to. That is the decision the rest of the product hangs off. An integration between six tools is six sets of failure modes. A single model has one. ## A payroll engine built from scratch The hardest part was payroll. No off-the-shelf payroll option fitted the agency tax model, so we built the payroll engine ourselves rather than bending the business to a provider that did not understand it. Building payroll is not a decision you take lightly: it is the part of the product where a bug costs someone their wages. Owning it means payroll reads from the same data model as the shifts and credentials, instead of from an export of them. ## Launched in ten weeks, then weekly releases The first version went live ten weeks after kickoff and replaced the six-tool stack for every pilot agency. Since then the product has shipped on a weekly release rhythm, which keeps each change small enough to reason about in a system that pays people. ## Kubernetes on AWS, chosen rather than defaulted to Our bias is towards the smallest infrastructure that meets the real availability requirement. On Oly Care that genuinely is Kubernetes on AWS, with PostgreSQL underneath and Stripe — not because it looks good on a diagram, but because the product needs it. On a retainer we live with our own architectural decisions rather than handing them over and leaving, which is a strong incentive to make them carefully. The DevOps and cloud work is ours as well as the application code. ## How the engagement runs Oly Care has a dedicated Brigada team: a named group working only on the product, in its standups and its tooling, as the engineering team of record on a multi-year retainer. Laravel and Vue 3 with Inertia on the application side. We work from Prishtinë on Central European Time, an hour ahead of the UK, which in practice is the whole working day in common. ## Where it stands Oly Care is live, replaced the six-tool stack for every pilot agency, and Brigada is now the engineering backbone of a growing SaaS business. Built and run by a dedicated Brigada team; the product is at oly.care. Q: What is Oly Care? A: Oly Care is a healthcare staffing platform for agencies: candidate credentialing, a shift marketplace, payroll and compliance in one product, replacing a stack of six separate tools. Q: Who built Oly Care? A: Brigada.dev, a software studio in Prishtinë, Kosovo. Brigada is the engineering team of record on a multi-year retainer, with a dedicated team on the product. Q: Why build a payroll engine instead of using a provider? A: No off-the-shelf payroll option fitted the agency tax model, so the payroll engine was built from scratch on the same data model as shifts and credentials. Q: What does Oly Care run on? A: Laravel, Vue 3 and Inertia, PostgreSQL and Stripe, deployed on AWS with Kubernetes. ======================================================================== # Kinsail — case study URL: https://brigada.dev/work/kinsail Industry: HR / Fintech Product: https://www.kinsail.co/ Stack: Laravel, Livewire, Tailwind, MySQL, Stripe, AWS UK workplace nursery scheme platform. Salary exchange, employer dashboard, and HMRC-compliant reporting engine built from scratch. Outcome: Launched the public-facing product and onboarded founding employers. Sign-ups convert without support calls. ## The problem: a financial product with no way to administer it The founder had the financial product. What did not exist was any way for an employer to actually administer it. HR teams needed something that did not require training, tax modelling, or a spreadsheet maintained by one person who understood it. A workplace nursery scheme is salary-exchange childcare that saves participating parents thousands a year. The saving is only real if the employer can run the scheme without a specialist. ## Minimum launchable product, in that order We went straight for the minimum launchable product and sequenced it deliberately: the salary-exchange calculator first as the entry point, employer dashboards second, automated reporting third. The calculator came first because it is the moment a parent understands what the scheme is worth to them. Everything after that is administration. ## Regulated UK payroll logic, built end to end Salary exchange touches payroll, and payroll answers to HMRC. The reporting engine was built HMRC-compliant from the start rather than retrofitted, because in this category a reporting bug is a compliance incident, not a ticket. Regulated UK payroll logic, built end to end by a team in Kosovo working UK hours. ## Sign-ups that convert without a support call The measurable goal was not traffic. It was whether an employer or a parent could get all the way through sign-up without picking up the phone. They can — which, for a product this regulated, was the actual objective. ## The stack Laravel and Livewire with Tailwind on the front, MySQL underneath, Stripe for payments, on AWS. Deliberately unexciting, because a regulated payroll product is not where you want to be debugging your framework choices. ## Where it stands The public-facing product shipped in 2024 and founding employers are onboarded. Built end to end by Brigada; the product is at kinsail.co. Q: What is Kinsail? A: Kinsail is a UK workplace nursery scheme platform: salary-exchange childcare that saves participating parents thousands a year, with an employer dashboard and HMRC-compliant reporting. Q: Who built Kinsail? A: Brigada.dev built it end to end — the parent-facing salary-exchange calculator, employer onboarding and dashboards, and the HMRC-compliant reporting engine. Q: Can a team in Kosovo build regulated UK software? A: Yes. Kinsail is regulated UK payroll logic built end to end from Prishtinë, on UK working hours. Central European Time gives a full working day of overlap with the UK. Q: What does Kinsail run on? A: Laravel and Livewire with Tailwind, MySQL, and Stripe, hosted on AWS. ======================================================================== # Avilog — case study URL: https://brigada.dev/work/avilog Industry: Aviation / SaaS Product: https://avilog.io/ Stack: Laravel, Nuxt, PostgreSQL, Redis, AWS, Offline-first mobile Aviation logbook platform for commercial and private pilots. Tracks flight hours, currency, licences, and crew rosters with one-click regulator exports. Outcome: Thousands of pilots, six years in production. We still run the engineering team. ## The problem: Excel, a 2009 app, and paper Pilots were logging flights in Excel, in an iOS app written in 2009, and on paper — often all three at once, because none of them did the whole job. Getting a regulator-ready export out of that took days of reconciliation. A logbook is not a nice-to-have for a working pilot. Currency, licences, and flight hours are what let you legally fly, and they have to be defensible to an authority that asks. ## Rebuilt around the regulatory schema Rather than digitising the paper logbook, we rebuilt it around the modern regulatory schema, with mobile-first data entry and automatic currency calculation. The data model matches what the regulator asks for, so the export is a read rather than a translation. ## Offline is the constraint, not a feature Pilots log flights where there is no signal, then export regulator-ready reports once they are back on the ground. Offline-first is not a bullet point on this product — it is the condition the entire data layer is designed around. ## One-click exports for EASA, FAA, and CAA Flight hours, currency, licences, and crew rosters all export in the formats EASA, FAA, and CAA expect. Three regulators, three formats, one source of truth underneath. ## Six years in production Avilog runs on AWS with PostgreSQL and Redis, and has been in continuous production for six years, serving thousands of pilots. Long enough that we have migrated it, upgraded it, and been paged by it — which is the only honest way to learn what an architecture actually costs. Brigada still runs the engineering team. That is the strongest evidence we can offer that people stay on projects here: the same relationship has outlasted most agency contracts several times over. ## Where it stands Six years live, thousands of pilots, and a long-running Brigada engineering team still shipping. The product is at avilog.io. Q: What is Avilog? A: Avilog is an aviation logbook platform for commercial and private pilots. It tracks flight hours, currency, licences, and crew rosters, and exports regulator-ready reports for EASA, FAA, and CAA. Q: Which regulators does Avilog export for? A: EASA, FAA, and CAA. The data model is built around the modern regulatory schema, so exports are one click rather than a manual reconciliation. Q: How long has Brigada worked on Avilog? A: Six years of continuous production, and Brigada still runs the engineering team. It is our longest-running client engagement. Q: What does Avilog run on? A: Laravel and Nuxt, with PostgreSQL and Redis on AWS, plus an offline-first mobile experience for logging flights without a signal. ======================================================================== # SEOMag Links — case study URL: https://brigada.dev/work/seomag-links Industry: Growth / SEO Product: https://seomag.ai/ Stack: Laravel, Inertia, React, PostgreSQL, Stripe Connect Premium link-building marketplace for SEO agencies. High-DR portals, sponsored placements, and automated order flow. Outcome: Live marketplace, automated order flow, and a growing agency customer base. ## The problem: links negotiated over email Agencies were negotiating placements manually over email and tracking them in spreadsheets, losing roughly 20% of the value to miscommunication — wrong anchor, wrong URL, placement that quietly disappeared a month later. Link building is a supply chain problem wearing a marketing costume. Everything that goes wrong with it goes wrong at the handoff. ## A two-sided marketplace with escrow We built it as a proper marketplace: escrow so neither side is exposed, automated outreach in place of the email thread, and verified placement tracking so a link that vanishes is visible rather than assumed. It launched with a curated catalogue of vetted high-DR portals, because a marketplace with no supply on day one is just a form. ## Payouts make or break a marketplace Stripe Connect handles payouts to portal owners. That is the part that quietly decides whether a two-sided marketplace survives: publishers who get paid predictably come back, and publishers who do not, do not. ## The stack Laravel with Inertia and React, PostgreSQL underneath, and Stripe Connect for the money movement. ## Where it stands The marketplace is live with automated order flow and a growing agency customer base. Built by Brigada. Q: What is SEOMag Links? A: A premium link-building marketplace for SEO agencies: high-DR portals, sponsored placements, escrow, and an automated order flow, with Stripe Connect payouts to portal owners. Q: Who built SEOMag Links? A: Brigada.dev, as product and engineering partner. Laravel with Inertia and React, PostgreSQL, and Stripe Connect. Q: Can Brigada build a two-sided marketplace? A: Yes. SEOMag Links is a working example: escrow, automated outreach, verified placement tracking, and split payouts to the supply side. ======================================================================== # Niche Data — case study URL: https://brigada.dev/work/niche Industry: Real Estate / Data Product: https://nichedata.ai/ Stack: Laravel, Inertia, Vue, PostgreSQL, Algolia Nationwide real-estate lead platform — daily-updated distress data, foreclosures, probates, and skiptracing, with CRM automation built in. Outcome: Nationwide daily-updated lead data, a thriving subscriber base, and infrastructure running a quarter of the planned footprint. ## The problem: a domain edge with no platform under it The founding team knew exactly which sellers were worth calling and why. What they did not have was a production-grade platform to deliver that knowledge to subscribers, or an infrastructure partner to run it. In lead generation the insight is cheap to describe and expensive to deliver. The value is in getting the right record, with the right contact details, in front of an investor before the next one sees it. ## What Niche Data is NicheOS connects distress data, CRM automation and coaching into one system, built to filter out tire-kickers so investors spend their time only on sellers who are ready to make a deal. The lists cover foreclosures, notices of default, probates, pre-probates, estate sales, divorces and guardianships, on top of parcel data — property details, tax records and ownership history — across more than 3,300 US counties. ## Search across the whole country Nationwide lead data is a search problem before it is anything else. Search runs through Algolia, because the data volume made ordinary database querying too slow to be useful to someone filtering a county for this week's probates. PostgreSQL holds the data and Algolia serves the search. Each does the one job it is good at, which is a better trade than asking one database to be both the record and the instant search box. ## From a record to a conversation A list is only worth something once you can reach the owner. Skiptracing attaches verified phone numbers, emails and relatives' contact records to a lead, and the built-in CRM takes it from there: leads upload straight in, follow-up is tracked, and text, email and direct mail campaigns run as automation rather than as a to-do list. Niche Data lists Twilio, SendGrid, Lob and Stripe among its integrations — the plumbing behind texts, emails, letters and billing. ## Small first, then grow The engagement started with a product-first sprint alongside the founder, then a lean full-stack build on Laravel, Inertia and Vue. We kept the team deliberately small at the start and grew it once the numbers justified it, rather than staffing up for a plan that had not met a customer yet. ## A quarter of the planned footprint The platform runs on infrastructure a quarter of the size of the footprint originally planned for it. That is the same bias we bring to every engagement: the smallest setup that meets the real requirement, with complexity added only when something forces it. ## Where it stands Niche Data is live nationwide with a thriving subscriber base, built by Brigada as product and engineering partner. The product is at nichedata.ai. Q: What is Niche Data? A: Niche Data is a US real-estate lead platform: nationwide distress lists such as foreclosures, probates and divorces, parcel data across more than 3,300 counties, skiptracing, CRM automation and coaching in one system called NicheOS. Q: Who built Niche Data? A: Brigada.dev, as product and engineering partner, starting with a product sprint alongside the founder and then a lean full-stack build. Q: Why does Niche Data use Algolia? A: The data volume made ordinary database querying too slow to be useful, so search runs through Algolia with PostgreSQL holding the data. Q: What does Niche Data run on? A: Laravel, Inertia and Vue, PostgreSQL, and Algolia for search. ======================================================================== # Vector Guest — case study URL: https://brigada.dev/work/vector-guest Industry: Hospitality / AI Product: https://vectorguest.com/ Stack: Laravel, LLM agents, RAG, Vector DB, LangFuse, Laravel Cloud The AI brain that runs aparthotel portfolios — revenue, demand, guest journey, and ground-team coordination through Harold OS. Outcome: A dedicated full-time Brigada team ships weekly with the founding operator, who opened up their entire stack to the platform. ## The problem: hospitality runs on shelf software Operators run on shelf PMS tools sold the same way to everyone. Data sits siloed across booking channels, email, and accounting. Teams grow to absorb repetitive manual work. The knowledge that actually runs the portfolio stays trapped in a handful of people's heads. Shelf software is sold identically to every operator, and each operator's real process gets wedged in around the edges. That gap — between how the software thinks the day runs and how the day actually runs — is where the manual work comes from. ## What Harold OS is Harold OS is a hospitality operating system: a system of records plus LLM agents that run a portfolio's entire operation — revenue, demand, guest journey, and every decision the ground team makes. In plain terms, it is the thing that prices the rooms, answers the guests, and runs the day. Not a feature bolted onto someone else's platform. The system itself. ## Forward-engineering instead of shelf software The approach is the opposite of generalise-first. Go deep into one operator's processes, own the solution end to end, and build one connected brain: tool-calling agents over a shared core that scales to the next operator without dragging the technical debt along. Harold OS is built operator-first, with a design partner running a multi-million-euro aparthotel portfolio. The founding operator opened up their entire stack — bookings, email, accounting, channel manager — to the platform, which is what makes agents useful rather than decorative. ## How the engagement runs A dedicated full-time Brigada team ships weekly with the founding operator. Same engineers from kickoff onward — not a rotating bench, and not a ticket queue with a spec written somewhere else. We work from Prishtinë on Central European Time, which is a full working day of overlap with the UK and EU, and afternoons with the US East Coast. In August 2026 the Vector Guest team — Harold, Matthieu, Erick and Andre — flew into the Prishtinë office for two days of product work. Whiteboards, a lot of coffee, and the kind of argument about a data model that never quite lands the same way over a video call. ## The stack Laravel on the back. Tool-calling LLM agents over a shared core. Retrieval-augmented generation against a vector database, so the agents answer from the operator's own records instead of guessing. LangFuse for tracing and evaluating what the agents actually did. Deployed on Laravel Cloud. Deliberately boring underneath the interesting part, because boring is what we can still debug at speed in year three. The model is rarely the hard part — evaluation is. ## Where it stands Harold OS is live, with a dedicated Brigada team shipping weekly alongside the founding operator. The product is at vectorguest.com. Q: What is Harold OS? A: Harold OS is the hospitality operating system Brigada builds for Vector Guest: a system of records plus LLM agents that run an aparthotel portfolio's entire operation — revenue, demand, guest journey, and the decisions the ground team makes. Q: Who builds Vector Guest? A: Brigada.dev, an IT outsourcing studio in Prishtinë, Kosovo. Brigada engineers the whole technology side as a dedicated full-time product team, shipping weekly with the founding operator. Q: What technology does Vector Guest run on? A: Laravel, tool-calling LLM agents, retrieval-augmented generation over a vector database, LangFuse for tracing and evaluating agent runs, deployed on Laravel Cloud. Q: Can Brigada build an AI product for a hospitality business? A: Yes. Brigada builds AI-native products end to end rather than adding AI features to shelf software — typically a two-week discovery sprint, a three-to-six month build, then a multi-year retainer with the same engineers. ======================================================================== # Prime Tech Group — case study URL: https://brigada.dev/work/prime-tech-group Industry: AI / Consumer Apps Product: https://oxodin.ch/ Stack: React Native, Expo, Laravel, LLM agents, Self-hosted TTS, Swiss hosting Swiss AI company. We build Oxodin — a voice-first agentic personal assistant on iOS, Android, and web, made in Switzerland. Outcome: Oxodin is live on the App Store and Google Play. Brigada continues as the engineering team. ## The brief: a digital employee, not a chatbot The vision was a 24/7 digital employee that accompanies you around the clock. You talk to it; it understands what you mean, manages your tasks and reminders, calls you back, and acts on your behalf. Voice-native, agentic, and running on Swiss infrastructure with end-to-end data sovereignty. No off-the-shelf assistant stack delivers all three at once, which is why it had to be built. ## What Oxodin does You record a voice message and Oxodin answers in audio. Tasks are created and managed by voice command. A memory and notes system keeps what you have told it, contacts sit alongside, and ChatOdin gives you a text chat with the same assistant, including file uploads. The product ships in English and German, with dark and light modes. ## One mobile codebase for iOS, Android and the web The mobile client is a single React Native and Expo codebase that ships to iOS, Android and as a progressive web app. The web application is Laravel with React and Inertia, and Laravel also sits behind the agent layer. One mobile codebase means one release to reason about across both stores. For a small product team shipping iteratively, that matters more than squeezing the last few percent out of a fully native build. ## Voice is the hard part People speak to Oxodin, so the agent layer has to feel instant. A pause that is acceptable in a chat window feels broken in a conversation. The voice-to-AI pipeline, the agentic intent architecture behind it, and push-notification-driven task loops — the part that lets the assistant come back to you — are where most of the engineering goes. ## Sovereignty as a constraint, not a slogan No data could leave Switzerland. That single requirement decided a lot: text-to-speech is self-hosted rather than sent to a third-party API, and the product runs on Swiss hosting. Self-hosting the voice is more work than calling someone else's endpoint. It is also the only way the promise on the box stays true. ## How the engagement runs Oxodin has been designed and shipped iteratively with the Swiss team. Brigada provides the product team — a lead developer, full-stack engineers and design — working from Prishtinë in the same time zone as Switzerland, so the whole working day is shared. ## Where it stands Oxodin is live on the App Store and Google Play, and Brigada continues as the engineering team. The product is at oxodin.ch. Q: What is Oxodin? A: Oxodin is a voice-first agentic personal assistant from Swiss AI company Prime Tech Group AG. It understands intent, manages tasks and reminders, calls you back, and acts on your behalf, on iOS, Android and web. Q: Who builds Oxodin? A: Brigada.dev is the product team behind Oxodin: a lead developer, full-stack engineers and design, working from Prishtinë, Kosovo with the Swiss team. Q: Why is the text-to-speech self-hosted? A: Because no data could leave Switzerland. Self-hosted TTS and Swiss hosting keep the voice pipeline inside the country. Q: What does Oxodin run on? A: React Native and Expo for the iOS, Android and web clients, Laravel with React and Inertia for the web application, LLM agents, and self-hosted text-to-speech on Swiss hosting. ======================================================================== # Walter — case study URL: https://brigada.dev/work/walter Industry: Sustainability / Web Product: https://walter-us.com/ Stack: Laravel 12, Statamic 6, Tailwind v4, MySQL, Forge, AI content pipeline US IT asset disposal and e-waste partner. We run their web platform — CMS, AI-assisted content ops, and lead generation. Outcome: The site is live and converts discovery calls; content publishing is automated and the platform runs on a maintenance retainer. ## Who WALTER is WALTER is a New York City based partner for IT asset disposal, office and facility clean-outs, product and packaging recovery, and debranding. It is an R2v3 certified partner and an ISO 9001, ISO 14001 and ISO 45001 certified operator, with data sanitisation aligned to NIST 800-88. It is also a social enterprise: the work is meant to create paid job pathways for young adults facing systemic barriers to employment. ## The problem: a credible platform a small team can run A growing operations company needed a web platform that matched how seriously it takes chain of custody: services, certifications, success stories and a steady stream of content. And it had to be run by a small, non-technical team, without a developer in the loop for every change. For a business that sells to compliance and procurement teams, the website is part of the due diligence. A buyer checking certifications at eleven at night should find them, current, without anyone having to send a PDF. ## Statamic on Laravel, with the content in the database We built it on Laravel 12 with Statamic 6 as the CMS and database-backed content, styled with Tailwind v4, on MySQL, deployed through Forge. Statamic gives the WALTER team a real editing experience; Laravel underneath means anything the CMS cannot do is ordinary application code rather than a plugin hunt. The editor supports rich text — bold, italics, lists and links — in the Insights articles, so the people writing the content control how it reads. ## A content pipeline that runs itself We added a scheduled, AI-powered article pipeline and automatic WebP image conversion, so content operations do not depend on someone remembering to publish or to resize an image. The AI here is quiet on purpose. It is a working part of lead generation — feeding the Insights section the site needs to be found — rather than a feature on a roadmap. ## Built around the discovery call Everything on the site points at one action: booking a discovery call. Around it sit the pages a buyer actually checks — certifications, success stories, client and team testimonials, news and press — plus a service wizard to find the right programme and an impact estimator. ## How the engagement runs After the build, the platform moved onto a maintenance retainer. We run it as well as having built it: updates, fixes and new content features go through the same team, working from Prishtinë with afternoons in common with the US East Coast. ## Where it stands The site is live and converts discovery calls, content publishing is automated, and the platform runs on a Brigada maintenance retainer. WALTER is at walter-us.com. Q: What is WALTER? A: WALTER is a New York City based IT asset disposal, e-waste and decommissioning partner, R2v3 certified, with a social mission of creating paid job pathways for young adults facing barriers to employment. Q: What did Brigada build for WALTER? A: The entire web platform: Laravel 12 with Statamic 6 as the CMS, database-backed content, a scheduled AI-assisted article pipeline, automatic WebP conversion, and lead generation built around booking a discovery call. Q: Can a non-technical team run a Statamic site? A: Yes. That was the brief on WALTER: services, certifications, success stories and articles are all edited by the WALTER team without a developer in the loop. Q: Does Brigada maintain sites after launch? A: Yes. WALTER runs on a Brigada maintenance retainer, so the team that built the platform also runs it. ======================================================================== # Helix Flag — case study URL: https://brigada.dev/work/helix-flag Industry: Fintech / Risk Product: https://helixflag.com/ Stack: Python, FastAPI, React, Chrome extension, Stripe, Sentry Collaborative B2B risk intelligence — businesses share and check red, yellow, and green risk flags to stop fraud before it happens. Outcome: Releases are gated and regression-tested, the risk engine catches flags it used to miss, and the team ships features instead of firefighting. ## What collaborative risk intelligence means Helix Flag lets businesses raise and check customer risk flags collaboratively — chargebacks, fraud, defaults — as red, yellow, and green flags across a network of over 3 million records. The value is in the network: one business flagging a bad actor protects the next. ## The problem: traction on fragile foundations The platform worked and had customers. Underneath, it had no CI gate, schema drift between the code and the database, a signup and billing flow duplicated in five separate places, and a browser extension that could return a different verdict from the web app on the same risk check. That last one is the dangerous kind of bug. A risk product that disagrees with itself is worse than no risk product, because someone acts on the answer. ## Audit first, then systematic hardening We ran a full audit before touching anything, then worked through it: CI hard gates so unreviewed code cannot ship, Sentry on both ends, end-to-end browser tests, and an operations runbook so the knowledge is not in one person's head. For the risk engine specifically, golden-record regression tests — a fixed set of records with known correct verdicts that every release has to reproduce exactly. ## One verdict engine, two surfaces The extension and the web app now share a single verdict engine, so they cannot disagree. Two surfaces, one answer, by construction rather than by discipline. ## The stack Python and FastAPI on the back, React on the front, a Chrome extension alongside it, Stripe for billing, and Sentry across both ends. Brigada works in Python where the problem is data or risk, not only in Laravel. ## Where it stands Releases are gated and regression-tested, the risk engine catches duplicate-record flags it used to miss, and the team ships features instead of firefighting. Hardened by Brigada; the product is at helixflag.com. Q: What is Helix Flag? A: Helix Flag is a collaborative B2B risk intelligence platform. Businesses raise and check red, yellow, and green risk flags on customers — chargebacks, fraud, defaults — across a network of over 3 million records. Q: What does platform hardening actually involve? A: On Helix Flag it meant a full audit first, then CI hard gates, Sentry on both ends, end-to-end browser tests, an operations runbook, golden-record regression tests for the risk engine, and a shared verdict engine so the browser extension and web app always agree. Q: Does Brigada take over existing codebases? A: Yes. Helix Flag came to us with traction and fragile foundations. We audit first, then harden systematically, so the team can ship features instead of firefighting. Q: Does Brigada work in Python? A: Yes. Helix Flag runs on Python and FastAPI with a React front end and a Chrome extension. We use Python where the work is data, risk, or AI. ======================================================================== # Street Wallet — case study URL: https://brigada.dev/work/street-wallet Industry: Fintech / Payments Product: https://www.streetwallet.co.za/ Stack: QR payments, Wallet integrations, Merchant onboarding, Next-day settlement Cashless QR payments for traders, small businesses, and corporates in South Africa — Apple Pay, Samsung Pay, SnapScan, and Zapper in one wallet. Outcome: The platform serves traders, small businesses, and corporate programmes across South Africa, with Brigada engineering ongoing since early 2026. ## The problem: card payments that exclude the informal economy Informal traders and small merchants were locked out of card infrastructure. Terminals are expensive, settlement is slow, and nothing on the market unified the local payment wallets their customers actually use. For a car guard or a street vendor, a card machine that costs money up front and pays out days later is not a payment solution. It is a reason to stay on cash. ## Three kinds of merchant, one product Street Wallet serves three groups. Traders, such as street vendors and car guards. Businesses, such as restaurants, salons and car washes. And corporates with many sites, such as fuel stations and hospitality chains. Each of them accepts payment by QR code, and the cash is available the next day — which matters when a business runs on its daily takings. ## One QR code, every wallet The customer should not have to care which wallet the merchant supports. Street Wallet pulls Apple Pay, Samsung Pay, SnapScan and Zapper into one wallet, so a single QR code takes payment from whichever app the customer already has. That is where the engineering weight sits: wallet integrations, the QR payment flows on top of them, and the merchant onboarding that gets a trader accepting payments in the first place. ## Tips, wages and feedback for corporate programmes For corporates, Street Wallet is a centralised way to handle tips, wages and marketing. Tips can be split from the bill, and co-branded QR codes collect tips and customer feedback while giving the business insight into its customers. ## How the engagement runs Street Wallet works with embedded engineers: Brigada people inside the existing team, reporting into Street Wallet's own leads, on a month-to-month arrangement. It is the lightest way to work with us, and it suits a company that already has a product and a direction and needs more engineering hands on it. South Africa runs on UTC+2, the same clock as Prishtinë in summer and an hour ahead in winter, so the working day overlaps almost completely. ## Where it stands The platform serves traders, small businesses and corporate programmes across South Africa, with embedded Brigada engineers on it since early 2026. Street Wallet is at streetwallet.co.za. Q: What is Street Wallet? A: Street Wallet is a cashless QR payments platform for South African traders, small businesses and corporates. It brings Apple Pay, Samsung Pay, SnapScan and Zapper into one wallet, with cash available the next day. Q: Who does Street Wallet serve? A: Traders such as street vendors and car guards, businesses such as restaurants, salons and car washes, and corporates with many sites, such as fuel stations and hospitality chains. Q: How does Brigada work with Street Wallet? A: As an embedded engineering partner since early 2026: Brigada engineers inside the Street Wallet team, reporting into its leads, on a month-to-month basis. Q: Can a team in Kosovo work with a company in South Africa? A: Yes. South Africa is on UTC+2, the same time as Prishtinë in summer and one hour ahead in winter, so the working day overlaps almost entirely. ======================================================================== # Evergreen Valleys — case study URL: https://brigada.dev/work/evergreen-valleys Industry: Real Estate Product: https://evergreenvalleys.com/ Stack: Laravel 11, Filament v3, Spatie Media Library, Tailwind, Vite Marketing platform for a sustainable residential development in Kosovo — 3D-rendered house models, bilingual, fully content-managed. Outcome: The site is live and drives booked appointments for the sales team; every model and page is editable without a developer. ## The problem: selling a neighbourhood people cannot walk yet A premium development needed a web presence that matched the product. That meant immersive visuals, content in both Albanian and English, and a way for a non-technical sales team to manage house models, locations and leads on their own. When the product is a house that is not finished, the website does a job a show home would normally do. It has to make the place feel real enough that someone books a visit. ## What we built: a digital showroom The site is built around atmospheric 3D renders and a house model explorer, with Model One and Model Two each on its own page. A neighbourhood plan shows the layout of the community, and the sustainability story — a neighbourhood designed to reduce its environmental impact — sits alongside the homes rather than in a footnote. Every route through the site ends at the same place: booking an appointment with the sales team. That is the primary call to action, because a visit is what sells a house. ## An admin the sales team actually runs The back office is a Filament v3 admin panel on Laravel 11. Houses, properties, gallery categories, the team page, per-page SEO, footer links and the legal pages are all records the sales team edits. Contact requests and newsletter sign-ups land in the same panel, so leads are not split across inboxes. The visual content is heavy, so Spatie Media Library handles it: renders and gallery images are attached to the record they belong to, instead of living in a folder someone has to remember. ## Bilingual from the data model up Albanian and English are both first-class. Translations are their own resource in the admin, so the sales team edits interface copy in both languages in the same place they edit the houses, and visitors switch between ALB and EN. Putting translations in the admin rather than in language files in the codebase is a small decision that removes a developer from every copy change. ## How the engagement ran Evergreen Valleys was a project build: a defined scope with a start and an end, covering design and build. It suits a first version or a specific platform, and the clear brief here — a showroom the sales team could run alone — made it a good fit. ## Where it stands The site is live and drives booked appointments for the sales team, and every model and page is editable without a developer — which was the entire brief. Designed and built by Brigada; the development is at evergreenvalleys.com. Q: What is Evergreen Valleys? A: Evergreen Valleys is a sustainable private residential community in Kosovo. Its website is a bilingual digital showroom with 3D-rendered house models and appointment booking. Q: Who built the Evergreen Valleys website? A: Brigada.dev designed and built it as a project build, from Prishtinë, Kosovo. Q: Can the sales team update the site without a developer? A: Yes. Houses, properties, galleries, SEO, translations and legal pages are all managed in a Filament v3 admin panel, and leads from the contact form land there too. Q: What does the Evergreen Valleys site run on? A: Laravel 11 with a Filament v3 admin, Spatie Media Library for the visual content, Tailwind and Vite. ======================================================================== # Rrini — case study URL: https://brigada.dev/work/rrini Industry: Travel / Marketplace Product: https://rrini.com/ Stack: Laravel 13, Inertia, React 19, SSR, Paysera, iCal channel sync Booking platform for short-stay apartments in the Albanian-speaking market, built for fast, last-minute bookings: instant booking, card payment, keyless check-in. Outcome: Live in Prishtina and Podujeva, with more Kosovo cities listed as coming soon. Brigada built it from the ground up; its founders are shareholders. ## A disclosure first: we own part of this one Since July 2026, Brigada's founders, Endrit Saiti and Ardit Xhaferi, have been shareholders in Rrini. Endrit stayed on as a shareholder and technology partner, advising on the product and its technical direction. That makes this case study different from the others on this page, so we are saying it at the top rather than in a footnote. Everything below describes the product as it is live at rrini.com and the code our team wrote. ## The brief: book tonight, pay by card, walk in without a key Rrini is built for the booking that happens late: someone who needs a place in Prishtina tonight, not in three weeks. So the product leads with three promises — instant booking, check-in without a physical key, and stays of three or six hours as well as by the night. Guests search by place, dates and number of guests, filter by the kind of stay they want — city apartments, villas, mountain stays, pet-friendly, family or couples — and pay securely online by card, with prices in euros. ## What hosts get Owners list their place, set their own prices and rules, and message guests directly. Rrini's pitch to hosts is that they keep the full amount rather than paying a 15% commission to a global platform. For owners who do not want to run it themselves, there is an optional management service: calendar sync with Booking.com and Airbnb, booking management across platforms, professional cleaning after each guest, tax declaration with ATK, and 24/7 guest communication. ## The hard part: money and calendars Two things decide whether a booking platform can be trusted: whether a payment confirms exactly one booking, and whether a night sold somewhere else stops being sold here. Both got more engineering than anything a guest sees. Payments run through Paysera, as a card checkout and as QR bank payments reconciled automatically against the wallet's statements. Fulfilment is idempotent in two layers — a row lock inside a transaction and a unique constraint on the transaction id — so a statement that arrives twice cannot confirm a booking twice. A single booking state machine owns every status change; nothing updates a booking's status directly. Calendars sync with Airbnb and Booking.com over iCal, so a night booked on another channel is blocked on Rrini. Guests get their check-in code and instructions, booking confirmations and reminders by SMS and email, which is what makes check-in without a physical key work. ## Two languages, and guardrails from real bugs Rrini runs in Albanian and English, server-rendered with Laravel 13, Inertia and React 19. A pre-push hook checks every push for parse errors, missing class references, undefined JavaScript variables, a client and SSR build that compiles — and any string that exists in only one language, so a guest never sees the wrong one. Those checks exist because the same kind of bug once shipped three times in one week: green build, broken page in a guest's browser. The fix was not more care. It was a check that makes the mistake impossible to push. ## Where it stands Rrini is live in Prishtina and Podujeva, with Prizren, Peja, Gjakova and Ferizaj listed as coming soon. It is operated by Rrini Solutions L.L.C. in Prishtinë, built by a Brigada team, with Endrit Saiti as technology partner. Q: What is Rrini? A: Rrini is a booking platform for short-stay apartments in the Albanian-speaking market, built for fast, last-minute bookings, with instant booking, online card payment, keyless check-in and three- or six-hour stays as well as nightly ones. Q: Who built Rrini? A: Brigada.dev built the platform from the ground up. Endrit Saiti, Brigada's co-founder and CTO, stayed on as a shareholder and technology partner, advising on the product and its technical direction. Q: Does Brigada have a stake in Rrini? A: Yes. Since July 2026, Brigada's founders, Endrit Saiti and Ardit Xhaferi, have been shareholders in Rrini. Q: Where does Rrini operate? A: In Kosovo, live in Prishtina and Podujeva, with Prizren, Peja, Gjakova and Ferizaj listed as coming soon. Q: What does Rrini run on? A: Laravel 13 with Inertia and React 19, server-rendered, Paysera for card and QR payments, and iCal calendar sync with Airbnb and Booking.com. ======================================================================== # Journal Field notes and hiring write-ups from the studio. ## we're hiring: mid laravel developer URL: https://brigada.dev/journal/hiring-mid-laravel-developer Published: 2026-09-30 · Hiring · 1 min read One seat, mid level, Laravel. You have shipped it to production and you would do it again. Real products for real clients, mostly in the US, UK and Europe. ## that's a wrap on future 2 tech 2026 URL: https://brigada.dev/journal/future-2-tech-2026 Published: 2026-09-28 · Field note · 3 min read Two days at Stand 36 in the Palace of Congresses, Tirana. What we came to say, who we met, and why a Prishtinë studio crossed the border to say it. ## built on standards URL: https://brigada.dev/journal/built-on-standards Published: 2026-09-18 · Field note · 3 min read We are ISO 9001 and ISO 27001 certified. What the audit actually asked for, what we had to change, and why a Kosovo studio bothered with the paperwork. ## one summer at brigada URL: https://brigada.dev/journal/one-summer-at-brigada Published: 2026-09-10 · Field note · 3 min read Four months, one game, and real code running for a real client. How ouroboros turned into an internship that distilled a voice model now sitting in a client's staging environment. ## circa, by haxhi bekaj URL: https://brigada.dev/journal/circa-by-haxhi-bekaj Published: 2026-08-31 · Field note · 1 min read Our friend and colleague Haxhi built a radial menu for macOS. Hit a hotkey, flick toward a slot, and the action fires. €9.99 once, no subscription. ## two days with vector guest URL: https://brigada.dev/journal/two-days-with-vector-guest Published: 2026-08-10 · Field note · 2 min read Harold, Matthieu, Erick and Andre flew into Prishtinë for two days of product work. Notes on why the flights were worth it — and on building a whole product instead of a slice of someone else's. ## ouroboros: the internship you win by out-thinking an AI URL: https://brigada.dev/journal/ouroboros Published: 2026-05-26 · Hiring · 3 min read Round two of how we hire. Three summer-internship seats, no code required — trick a bot into leaking a secret, then build the bot nobody can trick. ## we spoke at adc ulqin URL: https://brigada.dev/journal/adc-ulqin-2024 Published: 2024-05-18 · Field note · 2 min read The Albanian Developers Conference, third edition, inside the old fortress at Ulqin. We presented the company, demoed a programming book for Albanian-speaking children, and left copies in the museum archive. ## slug-or-die: how we hired this quarter URL: https://brigada.dev/journal/slug-or-die Published: 2026-05-13 · Hiring · 5 min read We deleted the recruiter funnel and replaced it with a public leaderboard — one bug, one judge, and a default-REJECT you had to talk past. Round one is over; here is how it went. ======================================================================== # Endrit Saiti, Co-Founder and CTO URL: https://brigada.dev/about/endrit-saiti Profiles: https://www.linkedin.com/in/endrit-saiti/ | https://github.com/endritvs Focus: Software architecture, Laravel, PHP, React, AWS, Kubernetes, Large language model systems, Retrieval augmented generation, DevOps Co-founder and CTO of Brigada.dev in Prishtina, Kosovo. Works on system architecture, Laravel, AWS infrastructure and production LLM systems for clients in the US, UK and Europe. I co-founded Brigada in 2024 with Ardit Xhaferi and I run engineering. In practice that means I am still in the code most days, which is deliberate. The moment a CTO stops reading the codebase they start making architecture decisions from memory, and memory is usually about two years out of date. Most of my work sits in three places. System architecture, where the decisions are expensive to reverse and worth arguing about. Infrastructure, mainly AWS, including the CJIS compliant environment we run for a US law enforcement client and the Kubernetes setup behind a UK healthcare platform. And production AI, which is the newest of the three and the one where the industry has the least settled practice. On AI I have a narrow opinion: the model is rarely the hard part, evaluation is. We have shipped LLM pipelines into environments where a wrong answer reaches a prosecutor, and the only thing that makes that survivable is a graded corpus you can run before every change. We build that first now, on every AI project, because we learned what happens when you do not. The longest running thing on my desk is Avilog, an aviation platform we have run for six years. Six years is long enough to be wrong about something and then have to live in it. That shapes how I scope work more than anything else I could put on a slide. ======================================================================== # Ardit Xhaferi, Co-Founder and CEO URL: https://brigada.dev/about/ardit-xhaferi Profiles: https://www.linkedin.com/in/ardit-xhaferi/ | https://github.com/ArditXhaferi Focus: IT outsourcing, Nearshore software development, Product strategy, Engineering team structure, Client partnership, Software delivery Co-founder and CEO of Brigada.dev in Prishtina, Kosovo. Works on product strategy, client partnerships and how outsourced engineering teams are structured for companies in the US, UK and Europe. I co-founded Brigada in 2024 with Endrit Saiti. He runs engineering, I run the company, and the line between those two is thinner than an org chart would suggest. My job is mostly the gap between what a client needs and what a team can actually deliver. That sounds soft until you have watched a project fail. Almost none of them fail for technical reasons. They fail because a decision took two weeks, because the scope quietly changed and nobody said so, or because the person paying for the work and the person using it wanted different things and never met. So I spend my time on the unglamorous side of delivery. Who the decision maker is. What is genuinely fixed and what is negotiable. Whether the team has enough context to make good calls without asking. Whether the client is hearing the truth from us when a date is at risk, which is the single thing that most determines whether they stay. On outsourcing I have a strong view, having been on both sides of it. The model works when the same people stay on your product long enough to care about it, and it fails when they are treated as fungible capacity. That is why we stayed small on purpose, and why we would rather turn work down than staff it badly. ======================================================================== # Contact Email: admin@brigada.dev Location: Prishtina, Kosovo (Europe/Pristina, CET) Website: https://brigada.dev