We build production AI systems for SMEs and scaling businesses - not demos, not proof-of-concepts that never ship. Agents that run in real workflows, RAG pipelines with actual retrieval quality, LLM integrations that hold up under load. We also use the same tools internally: our own coding assistants, evaluation harnesses, and automation pipelines. You would own both sides of that work.
The work
- Design and ship AI features into client products: agents, RAG, structured extraction, workflow automation
- Build and maintain RBB LAB's internal AI tooling and evaluation infrastructure
- Evaluate models, prompting strategies, and architectures against real production requirements - not benchmarks
- Work directly with clients to understand the problem before touching the stack
- Keep up with the field without chasing every new release
What we need
- Production experience shipping LLM-based systems - something running, not something demoed
- Strong fundamentals in at least one backend language (we mostly use Go and Rust, Python is fine for ML-adjacent work)
- A working understanding of retrieval, embeddings, and context engineering - not just API wrappers
- Comfort working in a small team where you own the decision, not just the implementation
- Based in San Marino or willing to relocate
The setup
Small team, senior-only. Compensation is competitive and discussed based on experience. Use the form below to apply - include your CV and a short note on a production AI system you built and what broke first.