Full-Stack Developer AI Products

To support our growth, we are looking for a new colleague (m/f)

Helion Research is bringing more of its technology development in-house. We are looking for a fullstack developer to help build the web applications our research teams use every day, including a growing set of AI-supported tools.

You will join a small internal development team and work closely with our technical lead, management and business users. Because the team is small, you get real responsibility early: your own components, direct contact with the people who use what you build, and a say in technical decisions. You won't be expected to figure everything out alone.

• Build features across our TypeScript stack: React frontends (Vite, TailwindCSS, TanStack Query) and typed backends (Hono, tRPC, Drizzle on PostgreSQL)
• Connect those applications to Helion systems, databases and external APIs
• Build AI features into the product, such as LLM-supported grading and validation of research data
• Help make AI-generated output reliable enough for production use
• Take prototypes to production: testing (Vitest, Playwright), documentation, code quality
• Try different technical approaches and measure what works You will work closely with the technical lead, and we expect you to progressively take ownership of your own components and projects.

Above all, a strong developer.

• Good software-development fundamentals
• Experience with full-stack, backend or web application development
• Solid TypeScript (we use it across the entire stack)
• React
• Databases and SQL (we run on PostgreSQL)
• Building or integrating APIs
• Git / GitHub
• Able to read existing code and build reliable new functionality on top of it
• Genuine interest in building with LLMs, shown through work projects, side projects, studies or open source
• Basic understanding of how LLM applications work: what inference costs in time and money, and how RAG fits together (embeddings, a vector store, retrieved context going into the prompt)
• Comfortable working primarily from our Antwerp office

Our platform already ships AI features: LLM-supported grading and validation of research data, and automated image processing. You will work on these from day one, so hands-on familiarity with LLM APIs, tool calling or AI coding tools (Claude Code, Cursor, Copilot) helps.

You should be able to build and debug a RAG pipeline from existing components, and reason about when a call to a model is worth its cost and latency. Training or fine-tuning models is not part of this job. Professional AI experience is a plus, not a requirement. If you have built something with an LLM, we would rather see that than a list of keywords on a CV.

• Parts of our stack: Bun, Vite, TailwindCSS, Hono, tRPC, Drizzle, Zod, TanStack Query
• Cloudflare Workers or other serverless platforms
• Testing with Vitest, Playwright or similar
• Docker
• Authentication, permissions and application security
• PDF or image processing

You like building, and would rather show a working prototype than discuss what could theoretically be built. You are curious about new tools and follow what is happening in AI development. You want to learn fast, working next to an experienced technical lead. You take ownership of problems once you understand them. And you care whether what you build actually solves someone's problem.

You understand the difference between an experiment, a prototype and a production application.

You join at the moment we are building up our internal development capability. You work directly with the technical lead, management and end users, and see your work used in real projects within weeks. There is room to experiment with new technologies, including the latest LLM tooling, and a clear growth path towards a more senior engineering or product role.

A positive & dynamic atmosphere in a fast-growing company, of driven professionals & change-makers. A full time, permanent position at our office in Antwerp, Belgium. A salary based on experience + extra-legal benefits + flexible work hours & the possibility of doing home office (2 days a week).

Please send us your CV & motivation