P

Polymath

Simulated worlds for end-to-end AI training

Member of Technical Staff - Research

AI ResearchConfirmed live in the last 24 hours
Full-timeSan Francisco, CA, USA$150k - $300k /yr

About the Job

Requirements

  • Have strong engineering and research fundamentals and be a prolific user of artificial intelligence tools.
  • Have experience post-training frontier models.
  • Have experience shipping reliable, production-quality code.
  • Have a track record of publications.

Responsibilities

  • Investigate failure modes of frontier models.
  • Create rigorous benchmarks to evaluate frontier agents on complex, realistic tasks requiring long-horizon reasoning and tool use in dynamic environments.
  • Post-train agents to reason, plan, and act over extended time horizons.
  • Publish research.
  • Build benchmarks, create environments, write production code, and run rigorous experiments.

About the Company

Polymath

Simulated worlds for end-to-end AI training

Polymath AI Labs builds simulated worlds where AI agents learn by doing real-world style tasks. Its simulations combine running applications, actual tools, and multi-step tasks so agents practice end-to-end workflows rather than isolated problems. The Horizon-SWE benchmark evaluates an agent’s ability to handle the full software development lifecycle—planning, coding, testing, and deployment—within a simulated software company. Polymath aims to enable autonomous AI that can tackle complex, long-horizon tasks with minimal human supervision across real-world settings.

Company size
1-10
Company stage
Seed
Total funding
$130K
Headquarters
San Francisco, CA
Founded
2025

Recently Posted Jobs

Explore more opportunities at Polymath

See more jobs