About the role
Design and optimise prompts, RAG architectures, and AI context systems that make enterprise AI reliable at scale.
You own the space between the model and the product β from retrieval strategy to evaluation frameworks that prove quality objectively.
This is a rare hybrid role: part engineer, part editor, part scientist. You'll write evals as readily as you write prompts, and you'll spar with our AI engineers about which knobs to turn first.
What we're looking for
- 2+ years working hands-on with LLM applications
- Strong written reasoning and a portfolio of prompt / eval work
- Comfort with Python and at least one vector DB