5 months ago
San Francisco, CA, USAMid Level
Responsibilities
- Build diverse, high-fidelity environments that test agents in realistic settings.
- Design complex tasks requiring long-horizon reasoning and tool use.
- Develop robust verifiers that reliably measure agent performance.
- Improve infrastructure and tooling for running, debugging, and improving environments.
- Work with the research team to identify failure modes and create new tasks and benchmarks.
Requirements
- Strong engineering fundamentals.
- Enjoyment of first-principles development and open-ended technical problem solving.
- High agency and a strong bias toward shipping.
- High quality standards and a focus on building robust systems.
- Ability to work effectively in a team and contribute to scientific progress.
Benefits
- Contract-based role with the opportunity to transition into a full-time position.
Categories
About Polymath
Polymath is a data lab focused on improving long-horizon agent capabilities and alignment.