Research Scientist, Applied Research

present of AI; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Chai Discovery in San Francisco is seeking researchers to advance AI for drug discovery, focusing on foundation models and practical evaluations. The role emphasizes transforming complex biology problems into robust machine-learning methods rather than biology theory alone.

You will develop data-generation strategies, assess model behavior with rigorous evaluations, and design post-training techniques that perform well on real-world tasks. Strong ML and quantitative skills are essential.

Qualifications

• Strong ML research background applied to large models.
• Experience evaluating model behavior and performance.
• Ability to translate messy biology into ML-methods.

Responsibilities

• Plan data-generation efforts for foundation models.
• Build evaluations that measure model behavior honestly.
• Design post-training approaches that work on practical tasks.
• Collaborate with pre-training researchers and engineers.

Skills

ML depth

Quantitative background

Tools

Structure prediction tools

Probability & statistics

Job description

AI & Data · San Francisco office · FullTime

The role centers on making existing foundation models work better for real drug discovery problems. A strong candidate plans focused data generation efforts, builds evaluations that honestly measure model behavior, and designs post-training approaches that hold up on practical tasks. Depth in machine learning plus a quantitative background matters more than a biology degree, and the team favors simple, sturdy methods over clever fixes for narrow cases.

How it fits the AI era

The work sits directly on frontier models for biochemical structure and interaction, so AI is the product rather than a side tool. Much of the judgment stays human: deciding which biological problems are worth formalizing, choosing what data to generate, and judging whether an evaluation reflects real usefulness. The role rewards people who can translate messy biology into methods borrowed from other fields.

How it builds AI experience

A person here would gain hands‑on practice in post‑training and evaluation of large models applied to proteins, antibodies, and small molecules, close to how those models reach partner drug programs. Building that background could mean publishing or contributing to work in protein or molecule modeling, getting fluent with structure prediction and design tools, and strengthening probability and statistics. Experience in large codebases helps, since results ship quickly alongside pre‑training researchers and engineers.

Worth showing

• ML research publications
• Protein and molecule modeling
• Structure prediction tools
• Probability and statistics

This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.

About Chai Discovery

Frontier AI foundation models and computer‑aided design suite to predict and reprogram biochemical molecules, per the company.