Co-Op, Reaction Prediction & Benchmarking

Your Impact at LILA

Develop and evaluate machine learning methods supporting small molecule design and synthesis. Working with public and internal chemistry datasets, the co-op will pursue a focused research project in either reaction prediction and benchmarking or molecular property prediction for medicinal chemistry optimization.

Potential projects include building benchmarks to assess reaction prediction models and reaction plausibility, or developing and fine-tuning molecular property predictors using project-specific data. The primary project will be selected based on the candidate’s interests, experience, and team priorities. Expected outputs include reproducible data and evaluation workflows, benchmark results, and research recommendations to inform future model development and integration.

What You'll Be Building

• Plan and execute a Co-Op project in conjunction with the Molecular Intelligence team

What You'll Need to Succeed

• Pursuing a graduate degree (MSc, PhD) in Chemistry, Computer Science, Physics, or similar

Bonus Points For

• Python (data processing, model training, and evaluation)
• Familiarity with cheminformatics libraries (RDKit or similar)
• Basic machine learning (supervised learning, model evaluation metrics)
• Familiarity with chemistry concepts through coursework or research experience
• Bonus: experience with molecular property prediction, multi-task learning, self-supervised pre-training, or model fine-tuning
• Bonus: familiarity with reaction data formats, reaction SMARTS, or reaction databases (e.g., Pistachio, Reaxys, SciFinder)
• Bonus: background or interest in organic or medicinal chemistry