onsite
Machine Learning Scientist - Agents for Small Molecule Drug Design - Roche
ML Engineer
Lead the design of autonomous, LLM‑driven agentic workflows that integrate ML models, physics‑based methods, and cheminformatics tools to accelerate small‑molecule drug discovery.
About the role
Key Responsibilities
- Design and implement autonomous agentic pipelines that orchestrate machine learning models, physics‑based simulations, and cheminformatics tools for small‑molecule drug design.
- Develop and fine‑tune large language models to generate hypotheses, interpret chemical data, and guide experimental workflows.
- Collaborate with medicinal chemists and computational scientists to validate model predictions and iterate on agent design.
- Integrate new data sources and model outputs into scalable, reproducible workflows using Python and modern ML frameworks.
- Document methodologies, publish findings, and contribute to internal knowledge bases and external publications.
Requirements
- PhD or equivalent experience in computational chemistry, machine learning, or related field.
- Strong programming skills in Python and experience with ML libraries (PyTorch, TensorFlow).
- Proficiency in large language model development and deployment.
- Experience with cheminformatics toolkits (RDKit, Open Babel) and physics‑based modeling.
- Excellent communication skills and ability to work cross‑functionally in a fast‑paced environment.
Skills
pythonmachine learning