periodic-labs
Research Scientist, Scaling RL
$225K–$350K
Menlo Park, CA +1 · Full-time · Posted Oct 1, 2026
- Location
- Menlo Park, CA
Montreal, Canada - Experience
- Not specified; the posting welcomes final-year PhD students and is open to offering a 1-year post-doc.
- Education
- Bachelor’s degree or similar experience
- Visa sponsorship
- Visa sponsorship is available.
About the job
Research how reinforcement learning scales for scientific tasks, develop improved algorithms, and move findings from controlled experiments to large-scale training runs.
What you'll do
- Design experiments on how RL performance scales with compute, model size, data, and reward quality.
- Develop RL algorithms and adaptive sampling or curriculum methods for long-horizon tasks.
- Study RL training bias and stability, and improve training and inference compute efficiency.
What they ask for
- Hands-on experience training LLMs with reinforcement learning.
- Strong attention to detail and a rigorous scientific approach.
- Ability to develop small-scale RL setups that transfer to large-scale training runs.
- Comfort implementing, debugging, and testing research ideas across a complex training stack.
Pay
$225,000-$350,000 USD base salary
Tools
LLMs, reinforcement learning
AI focus
Train frontier models to develop scientific knowledge and reasoning, and research RL algorithms and scaling for scientific tasks.
Model-written summary · the original posting is authoritative