Hybrid AI Scientist for Drug Discovery & Model Training

BioSpace; San Francisco (CA); Hybrid; USD 168,000 - 268,000; Full time

Benefits offered by this job

401(k)

Pension

Medical benefits

Vacation benefits

Job summary

Lilly is seeking an Advisor - AI Scientist to design, train and evaluate foundation models advancing drug discovery in chemistry and biology. You will build generative and predictive models using Lilly data, bridging ML, chemistry, biology and physics to transform scientific insights into medicines.

This role blends rigorous research with practical impact. Based in Silicon Valley with a flexible hybrid model (three days onsite, two remote), you will collaborate across disciplines and publish or

Qualifications

• Deep expertise in computational chemistry/biology, genomics, or related field.
• Significant contributions to generative, predictive, or foundation models for science.
• Publications, patents, open-source work or notable scientific achievements.

Responsibilities

• Design, train, and evaluate foundation models for chemistry and biology.
• Develop generative/predictive AI models for molecular and biological systems.
• Translate ML research into scientific impact across discovery challenges.
• Collaborate with researchers, engineers and domain experts to accelerate AI in discovery.
• Lead research from concept to validated impact with practical applications.

Skills

Genomics

ML research

Foundation models

Python

GPU computing

Cross-disciplinary collaboration

Scientific communication

Industrial research experience

Education

PhD in Machine Learning or related field

Tools

PyTorch

JAX

Job description

Lilly is seeking an Advisor - AI Scientist to design, train and evaluate foundation models advancing drug discovery in chemistry and biology. You will build generative and predictive models using Lilly data, bridging ML, chemistry, biology and physics to transform scientific insights into medicines.

This role blends rigorous research with practical impact. Based in Silicon Valley with a flexible hybrid model (three days onsite, two remote), you will collaborate across disciplines and publish or