Actively Recruiting
Software Engineer - RL Environments
About Polymath Fellows
Polymath Fellows is a community of exceptional engineers and scientists focused on creating the environments and data needed to train the next generation of AI models. Fellows identify where current models fail, and create high-quality training environments to teach AI how to improve on realistic engineering and research workflows.
Our mission is to accelerate progress on the hardest problems in software engineering, biology, medicine, materials science, energy, and beyond. We work with the leading model labs to push the frontier of AI capabilities. Polymath is backed by Base10, Founders Future, Y Combinator, and other incredible investors & angels.
About the role
In this role, you’ll design and build reinforcement learning environments and tasks that challenge frontier coding models:
- Design and build RL environments and tasks that reflect real software engineering and systems work across domains like compilers, databases, distributed systems, networking, OS, security, and ML/GPU infrastructure
- Specify clear success criteria, failure modes, and grading signals for agent attempts
- Collaborate with Polymath researchers and other domain experts
Requirements
- Substantial professional or open-source experience in software engineering and the software development lifecycle
- Comfortable reasoning at the systems level (concurrency, memory, I/O, networking, performance, etc) and how things fail
- Comfortable in a containerized, reproducible dev loop with automated testing (or able to pick it up quickly)
- Systems engineering mindset: hands-on experience with Docker, Kubernetes, and low-level systems design
- Interest in how frontier coding models behave, and building environments and tasks that expose their shortcomings
Fellows must have expertise in at least one of the following areas:
- Deep understanding of performance engineering or distributed systems
- Background in cybersecurity
- Networking, proxies & service mesh
- Operating systems, virtualization & low-level/embedded
- DevOps or SRE experience running production systems at scale
- ML engineering experience
- Expertise with operating systems internals
Compensation
- US: $50-$110/hr
- India: $20 - $50/hr
- Canada / UK / Western Europe: $40 - $90/hr
- Eastern Europe / Latin America: $25 - $65/hr