← All roles

Research Scientist, Scaling RL

Periodic Labs· Engineering
Apply on company site
Menlo Park, CA; Montreal, Canada$225k - $350k + equityFull-timePosted 2w ago

About the role

Research scientist focused on scaling reinforcement learning for frontier models, exploring how RL scales with compute, model size, data, and reward quality; developing algorithms and conducting controlled experiments to apply findings to large-scale runs.

Responsibilities

  • Design experiments to understand how RL performance scales with compute, model size, data, and reward quality, building on work such as ScaleRL
  • Develop better RL algorithms, spanning policy optimization, advantage estimation, exploration, and credit assignment for long-horizon RL tasks
  • Build adaptive sampling and curriculum methods that adjust task difficulty, problem selection, and the number of rollouts as models improve
  • Study bias and stability during RL training, including importance-sampling corrections and methods to tackle policy staleness and training–inference mismatch
  • Improve compute efficiency across training and inference through experiments with hyperparameters, such as length penalties, rollout counts, batch sizes, and update schedules

Qualifications

  • Hands-on experience training LLMs with reinforcement learning
  • Strong attention to detail and rigorous approach to answer questions scientifically
  • Ability to come up with small-scale RL setups that transfer to large-scale training runs
  • Comfort working across a complex training stack to implement, debug, and test new research ideas
  • Minimum education: Bachelor’s degree or similar experience

Skills

Required:Reinforcement learning for LLMsExperiment design and analysis
Nice to have:Adaptive sampling and curriculum methodsPolicy optimization / exploration / credit assignmentTraining stack debugging and implementation

Benefits

  • Not specified in the description
Last updated Oct 21, 2026. Interested? Apply on the company site.

About Periodic Labs

From bits to atoms.

Periodic Labs is an artificial intelligence research company pioneering the next leap in scientific discovery. We operate at the intersection of machine learning and scientific experimentation under the thesis that that true understanding of the universe is only possible by interacting and experimenting on it.

artificial intelligencescientific discoverymachine learningscientific experimentation
Company profileWebsite

More open roles at Periodic Labs