Research Engineer

Barrington James; San Francisco (CA); On-site; USD 140,000 - 210,000; Full time

Job summary

Barrington James is seeking a Reinforcement Learning Engineer for a fast-growing AI company in San Francisco, at the intersection of frontier AI, healthcare and drug discovery. The role focuses on building RL environments and post-training systems for healthcare AI across the full lifecycle.

You will design environments, rewards and evaluation systems, run experiments, train models and agents, and improve data and feedback signals that models learn from.

Qualifications

• 2+ years of experience in reinforcement learning, post-training, agents or related ML systems.
• Experience with RL, RLHF/RLAIF, RLVR, reward modelling, agent environments or LLM post-training.
• Strong Python and ML engineering skills.
• Experience building experiments, evaluation frameworks and training pipelines.
• Strong understanding of data quality, dataset design and model evaluation.
• Ability to work with large ML codebases and debug training runs, data pipelines and model behaviour.
• A strong research mindset combined with the ability to move quickly from concept to working system.
• ICML, ICLR or NeurIPS publications are highly desirable.

Responsibilities

• Design reinforcement learning environments, reward structures and evaluation systems across the RL lifecycle.
• Run experiments, train models and agents, and analyze results.
• Identify failures and iteratively improve data, feedback signals and model behaviour.

Skills

Reinforcement learnin

Python

ML engineering

Experiment design

Training pipelines

Data quality

Debug ML codebases

Research mindset

Publications desirable

Job description

I’m currently supporting a fast-growing AI company working at the intersection of frontier AI, healthcare and drug discovery, and they’re looking for an RL Engineer to join their team in San Francisco.

This is a highly hands-on role focused on building reinforcement learning environments and post-training systems for healthcare AI. You’ll work across the full RL lifecycle, designing environments, rewards and evaluation systems, running experiments, training models and agents, analysing failures and continually improving the data and feedback signals that models learn from.

• 2+ years of experience in reinforcement learning, post-training, agents or related ML systems
• Experience with areas such as RL, RLHF/RLAIF, RLVR, reward modelling, agent environments or LLM post-training
• Strong Python and ML engineering skills
• Experience building experiments, evaluation frameworks and training pipelines
• Strong understanding of data quality, dataset design and model evaluation
• Ability to work with large ML codebases and debug training runs, data pipelines and model behaviour
• A strong research mindset combined with the ability to move quickly from concept to working system
• ICML, ICLR or NeurIPS publications are highly desirable

What matters most is strong RL/ML expertise, excellent technical judgment and an interest in applying frontier AI techniques to healthcare and biomedical problems.

If you're an RL/ML engineer excited by the opportunity to work on challenging problems with real-world impact, I'd be interested in hearing from you.