2 months ago
London, United KingdomSenior
Responsibilities
- Build RL environments across finance, healthcare, and legal domains.
- Design tasks with golden answers, calibrated rubrics, and programmatic reward signals.
- Write verification logic and reward functions that distinguish good model outputs from bad ones.
- Work with investment analysts, physicians, and attorneys to translate professional workflows into structured tasks.
- Prototype approaches to evaluation, verification, and synthetic data generation.
- Own problems end to end, from domain-expert scoping through delivery of a working environment.
Requirements
- Practical experience building working systems with LLMs, including prompting, evaluation, and agentic harnesses.
- Technical sharpness, high agency, and the ability to work independently without detailed specifications or roadmaps.
- Ability to work across engineering, evaluation design, and collaboration with domain experts.
- Demonstrated ability to ship, iterate, and improve working solutions.
- Experience using LLMs as part of the development process.
- Domain knowledge in finance, healthcare, or legal is a plus.
- Familiarity with RL concepts, model training, and post-training workflows is a plus.
- Cloud infrastructure experience with AWS or GCP is a plus.
- Previous startup experience, especially as an early engineer, is a plus.
Tech Stack
Categories
