About the RoleWe are looking for CADD / Application Scientists who want to help define the next generation of computational drug discovery tools, not just operate the current one.
You will be a scientific design partner for our model and training teams, bringing real discovery program experience into what we train on, what model behaviors matter, and which applications are worth building toward. As the models mature, you will also help bring these tools to pharma and biotech partners, shaping collaborations around problems where better molecular simulation could change real program decisions.
What You'll Do- Shape training data strategy for our models: identify which experimental, structural, partner-accessible, and synthetic data sources are likely to improve affinity prediction, selectivity, generalization, and downstream drug discovery utility.
- Own data and structure curation for high-value protein-ligand systems end-to-end: connect affinity measurements to assay context, prepare structures, assign protein and ligand states, review or generate poses, and label assumptions and uncertainty.
- Work with model and training teams to interpret model successes and failures, separating data problems, setup problems, and model limitations.
- Help decide which applications are worth pursuing, from lead optimization and selectivity to pose assessment, scaffold transfer, affinity prediction, hit rescoring, and future extensions beyond potency.
- Shape partner programs with BD and leadership, translating model capabilities into scientifically credible collaborations with pharma and biotech teams.
About You- You have significant experience in CADD, structure-based drug design, computational chemistry, medicinal chemistry collaboration, or related work in drug discovery.
- You have experience supporting, or leading external collaborations with pharma, biotech, or discovery partners.
- You have strong intuition for protein-ligand binding, ligand poses, assay artifacts, protonation / tautomer states, waters, cofactors, ligand strain, and where modeling workflows quietly go wrong.
- You are comfortable making expert judgments from imperfect data and can tell which benchmarks, application ideas, or partner case studies would matter to a real discovery team.
- You are excited to work closely with ML researchers, simulation scientists, and platform teams.
Nice to HaveEven if you hit none of these bonus features, we encourage you to apply.
- Experience curating and running protein-ligand affinity or FEP benchmarks, including OpenFE or related evaluation efforts.
- Experience across the full discovery arc from target identification through hit finding, hit-to-lead, and lead optimization.
- Experience designing scientific case studies, technical reports, or partner-facing demonstrations for new computational methods.
- Familiarity with ML-assisted drug discovery, active learning, synthetic data generation, or model evaluation for molecular systems.
