Prime Intellect
Research Engineer / Scientist - Reinforcement Learning
$350K
San Francisco +1 · Full-time · Posted Oct 16, 2026
- Location
- San Francisco
Remote - Visa sponsorship
- Visa sponsorship and relocation assistance for international candidates.
About the job
Develop reinforcement-learning algorithms and systems for training Prime Intellect’s open models, focusing on RL research, scaling experiments, or training frameworks.
What you'll do
- Develop and test RL methods for reasoning, coding, and long-horizon tasks; investigate training stability.
- Design controlled scaling experiments and turn findings into training recipes, validating results on larger models and longer runs.
- Extend prime-rl and improve performance, numerical correctness, and diagnostics across trainers, rollout workers, and environments.
What they ask for
- Demonstrated depth in RL algorithms, empirical research on model training, or distributed ML systems.
- Strong programming skills in Python and experience with PyTorch or a comparable ML framework.
- Ability to design controlled experiments, interpret results, and debug unexpected model or system behavior.
- Experience implementing, testing, and evaluating a research idea or systems change.
Nice to have
- Experience with distributed training frameworks or asynchronous rollout pipelines, or diagnosing unstable RL runs.
- Work on coding agents, verifiable rewards, or long-horizon environments.
Pay
$150-350k (USD; cash compensation, including equity incentives)
Tools
Python, PyTorch, prime-rl
AI focus
Research and build reinforcement-learning methods, scaling approaches, and frameworks to train open models for reasoning, coding, and long-horizon tasks.
Model-written summary · the original posting is authoritative