Senior GenAI Developer

Build AI systems that make sense in the real world.

Most companies are still trying to understand what AI adoption actually means. They run PoCs, connect a few APIs, build a chatbot, and hope the technology will carry the rest.

At Attrecto, we want to do something more serious.

We are a 70+ person European software company with big ambitions, strong engineering roots, and a growing AI R&D direction. We are not trying to compete with Big Tech job ads or hire someone who has already built foundation models at global scale. We are looking for a strong medior or senior AI developer who wants ownership, room to think, and the chance to shape how practical AI systems are built from the ground up.

This is a greenfield role in a team where AI is treated as an engineering discipline, not as a magic layer on top of software. We care about model behavior, orchestration, agentic workflows, data quality, security, compliance, user experience, and business fit. We also care about knowing when not to use AI.

The right person for this role understands that the best solution is not always the newest model, the biggest context window, or the most expensive API call. Sometimes the better answer is a smaller model, a cleaner architecture, a stronger retrieval strategy, a rules-based workflow, a sharper evaluation loop, or a simpler product decision.

If you think this way, we should talk!

What you will help build

You will work on R&D-driven Generative AI solutions that move beyond demos and into usable systems. This may include agentic workflows, RAG-based knowledge systems, AI-assisted internal tools, model integrations, rules-based automations, evaluation pipelines, and AI features embedded into client products.

You will not just “add AI” to applications. You will help decide how AI should behave inside a system: what the model should do, what it should never do, what should be deterministic, what needs human approval, what data can be used, how quality should be measured, and how the solution can be maintained after the first successful demo.

You will work closely with developers, PMs, designers, analysts, and business stakeholders. A large part of the role is understanding the real need behind a request, translating uncertainty into a workable architecture, and helping the team make smart technical decisions.

The kind of thinking we are looking for

We are looking for someone who understands agentic AI beyond the buzzword. You know that agents are not just “LLM calls in a loop.” You can reason about tools, state, memory, retrieval, planning, orchestration, permissions, guardrails, fallback paths, and failure modes.

You have hands-on experience with LLM-based systems and you understand how models behave. Prompting is not dead, but it is not the whole story either. You know how to guide model output, evaluate quality, spot hallucination risks, compare model capabilities, and design around uncertainty.

You have worked with RAG, embeddings, vector databases, structured and unstructured data, chunking strategies, retrieval quality, and evaluation. You understand that a knowledge base is only as good as its data, architecture, and feedback loop.

You can make informed model choices. You follow the AI ecosystem, but you do not confuse hype with value. You can explain why a specific use case needs GPT-4.1, Claude, Gemini, a smaller open model, a local deployment, a hybrid approach, or no LLM at all.

You care about the business case. You want to understand why the system is needed, who will use it, what success looks like, what risks exist, and how the solution fits into the customer’s workflow.

You treat data privacy, security, and compliance as part of the design. GDPR, EU AI Act considerations, sensitive data handling, access control, auditability, and human-in-the-loop patterns should not be afterthoughts.

What we expect from you

You should be strong in Python and comfortable designing software that other engineers can understand, extend, test, and operate.

You should have practical experience with Generative AI systems, including LLM APIs, RAG pipelines, agentic workflows, model orchestration, evaluation, and integration with external tools, APIs, databases, or enterprise systems.

You should be familiar with parts of the modern AI engineering stack, such as LangChain, LangGraph, Azure AI services, OpenAI APIs, Claude, Gemini, vector databases, graph databases, cloud platforms, CI/CD, monitoring, and deployment patterns. We do not expect you to know every tool. We do expect you to understand the underlying concepts well enough to choose responsibly.

You should be able to work in an environment where not everything is predefined. This role needs someone who can explore, challenge ideas professionally, propose options, validate assumptions, document decisions, and move from ambiguity toward working software.

You should be able to communicate clearly in English and work well with both technical and non-technical people.

What makes you stand out

You have built AI agents that interact with real tools, files, APIs, databases, or internal systems.

You have experience turning PoCs into production-grade solutions with versioning, documentation, evaluation, monitoring, and maintainability in mind.

You have worked with enterprise data, compliance-heavy environments, on-prem or hybrid deployments, or client-facing software projects where the requirements had to be shaped through discovery.

You have a strong opinion about where AI is useful, where it is risky, and where people are currently overengineering it.

Why join Attrecto now

Because this is still early.

You will join at a stage where your thinking can influence our AI standards, technical patterns, internal tools, and long-term direction. You will not be a small piece of a large machine. You will help build the machine with us.

We are a small company compared to global tech giants, but that is also the opportunity. Decisions are closer. Ideas travel faster. Ownership is real. If you see a better way to solve something, you can bring it to the table.

You will work in a supportive, people-first, geek-friendly company culture with flexible hybrid or remote work from Budapest, Győr, or elsewhere in Hungary.

We offer training, education, career development through our Growth Program, community events, team-buildings, sports activities, and a balanced work environment.

But the real reason to join is this: we want to build meaningful AI products, not just impressive prototypes.

If you want to help define how that should be done, let’s talk.