Research Engineer – Reinforcement Learning

This is a fully remote position, open to applicants in United States.

📋 Description

• Lead and advance a research agenda focused on enhancing model capabilities for long-term legal reasoning and agentic workflows.

• Design and conduct post-training experiments utilizing SFT and RL with top-tier open-weight models.

• Create reward functions and verifiers for semi-verifiable tasks, which include rubrics, learned preference models, and expert preference data.

• Investigate the limitations of reward functions and verifiers.

• Execute and troubleshoot large-scale training, addressing issues like divergence, reward hacking, and instability.

• Develop and utilize stringent evaluations to monitor progress and determine deployment decisions.

• Design, implement, test, and debug code throughout the research stack.

• Transform research concepts from experiments into training runs, leading to models employed by clients.

• Take responsibility for post-training methodologies across both supervised fine-tuning and reinforcement learning.

⛳️ Requirements

• Practical experience with post-training LLMs utilizing SFT and RL, including preference optimization.

• Direct accountability for training runs and their outcomes.

• Strong foundation in machine learning principles.

• Proven experience in ensuring reliable distributed training runs.

• Proficient engineering skills in Python.

• Ability to produce high-quality code and manage end-to-end system responsibilities.

• Research background in reinforcement learning for language models or reward modeling.

• Capability to swiftly transition from experiments to implemented improvements.

• Publications are encouraged but not mandatory.

• Must possess legal authorization to work in the United States.

• Must be eligible to work in the U.S. or obtain visa sponsorship as required.

🏝️ Benefits

• Equity plan.

• Comprehensive health coverage.

• Dental coverage.

• Vision coverage.

• Retirement benefits with a 401(k) match of up to 4%.

• Flexible paid time off (PTO).

• Equal opportunity employer.

• Reasonable accommodations available for applicants with disabilities.