AI Scientist-Engineer for Autonomous Molecule Discovery

BioSpace; San Francisco (CA); On-site; USD 152,000 - 222,000; Full time

Benefits offered by this job

Company bonus

Comprehensive benefits package

Job summary

Lilly is seeking a scientist-engineer hybrid to design the learning layer of our scientific agent platform. You will design the environments, rewards, and domain-specific models that enable agents to improve based on experimental feedback.

You'll translate wet-lab and computational endpoints into a trainable signal to build models that plan and act against them. You'll design the learning layer of the DMTA cycle, infusing automation with foundation models, multi-agent systems, and robotics to

Qualifications

• PhD (or MS + 3 yrs / BS + 5 yrs equivalent experience) in Machine Learning, Bioinformatics, Cheminformatics, Computer Science, or related STEM field with demonstrated wet-lab collaboration or hands-on experience.
• Approximately 1-2 years of demonstrated experience in applying AI/ML in scientific disciplines such as biology, chemistry, neuroscience, or a related field (industry postdoc counts)
• Hands-on experience training or post-training AI models

Responsibilities

• Design RL environments for real discovery tasks.
• Curate reward functions from noisy scientific signal.
• Post-train domain models (SFT, PPO, GRPO, reward modeling) on chemistry and biology tasks.
• Integrate learned policies with domain tools (RDKit, LIMS APIs, instrument drivers).
• Build the eval infrastructure: task suites, scoring harnesses, regression tracking, and experiment tracking.

Skills

Python

ML frameworks

Reinforcement Learning

Cloud computing

Mentoring

Education

PhD in ML/CS/Bioinformatics

Tools

RDKit

Molecular graph ML

ELN/LIMS APIs

Agent SDK

Kubernetes

Job description

Lilly is seeking a scientist-engineer hybrid to design the learning layer of our scientific agent platform. You will design the environments, rewards, and domain-specific models that enable agents to improve based on experimental feedback.

You'll translate wet-lab and computational endpoints into a trainable signal to build models that plan and act against them. You'll design the learning layer of the DMTA cycle, infusing automation with foundation models, multi-agent systems, and robotics to