We're building AI-native products. You'll work directly with our engineering team building and shipping RAG pipelines, agentic systems, and LLM-powered features — real production work.
What you'll actually do
- Build and iterate on RAG pipelines (retrieval, chunking, embeddings, eval loops)
- Design and ship agentic workflows — tool use, multi-step loops, error handling — not one-shot chat interactions
- Work inside a structured dev cycle (TDD or similar), writing evals before shipping, not vibes-checking outputs
- Debug, evaluate, and improve LLM-based systems in production
You must have
- Hands-on experience building RAG or LLM-based systems — side projects, hackathons, internships,
or work experience all count, as long as it's real and you can walk us through it
- Comfort working agent-first: tools, loops, evaluation frameworks — not just chatting with a model and pasting the output
- Familiarity with a real engineering workflow (structured dev cycles, testing discipline) — you ship code that's meant to survive contact with production
This is not for you if
- Your day-to-day AI usage is Copilot autocomplete or copy-pasting into ChatGPT
- You've only read about RAG/agents but haven't built one
- You want a slow-paced, process-heavy workplace