Research Engineer / Scientist, Reinforcement Learning

Harvey

6d • $230,000 - $340,000

About The Position

As a Member of Technical Staff, Modeling in Harvey Research, you will own how Harvey adapts and improves leading open-weight models for legal work. Following the post-training of its first model built for legal work, Harvey is building a team to scale its post-training efforts. You will own the post-training recipe across supervised fine-tuning (SFT) and reinforcement learning (RL), including reward design and verification for challenging, semi-verifiable legal tasks.

You will help set research direction and carry ideas from experiments through training runs to models used by customers. You will join a small, high-bar team of hands-on engineers and researchers, with each person owning a significant part of the work. Harvey is seeking research engineers and scientists who have run training at scale, are immersed in RL for language models, and want their work to reach users quickly.

Requirements

• Hands-on experience post-training LLMs with SFT and RL, including preference optimization, and direct ownership of training runs and results.
• Deep machine learning fundamentals and experience getting distributed training runs working reliably.
• Strong engineering skills in Python, with production-quality code and end-to-end system ownership.
• Research experience in RL for language models or reward modeling, with the ability to move quickly from experiments to shipped improvements.

Nice To Haves

• Publications are welcome, but not required.

Responsibilities

• Own and pursue a research agenda to improve model capability on long-horizon legal reasoning and agentic workflows.
• Design and run post-training experiments across SFT and RL, building on leading open-weight models.
• Develop reward functions and verifiers for semi-verifiable tasks, including rubrics, learned preference models, and expert preference data, and study where they break.
• Run and debug training at scale, addressing divergence, reward hacking, and instability.
• Build and use rigorous evaluations to track improvements and decide what ships.
• Design, implement, test, and debug code across the research stack.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed