Research Engineer, Reinforcement Learning

Job Summary

Owning the post-training agenda for legal models, the full-time Research Engineer, Reinforcement Learning will design and run experiments, develop reward functions, and manage training at scale in a remote environment.

Key responsibilities

• Own and pursue a research agenda to enhance model capabilities for long-horizon legal reasoning and workflows
• Design and execute post-training experiments utilizing leading open-weight models
• Develop reward functions and verifiers for semi-verifiable tasks, ensuring robust evaluation metrics

Required qualifications

• Hands-on experience with post-training LLMs using SFT and RL, including preference optimization
• Strong foundation in machine learning fundamentals and experience with distributed training
• Proficiency in Python, with a focus on production-quality code and system ownership
• Research experience in reinforcement learning for language models or reward modeling
• Ability to transition quickly from experiments to implemented improvements