AI Biologist - Structure Biology
LatchBio
Summary
You will be the scientific authority on ground truth, testing whether AI agents can reason like an expert structural biologist. Separating what the data actually support from what the model asserts is the core of the job. You will work with software engineers and scientists to build benchmark tasks and datasets across viral and bacterial proteins, toxins, RNA elements, and host-pathogen complexes.
Job description
About the Role
You will be the scientific authority on ground truth, testing whether AI agents can reason like an expert structural biologist. Separating what the data actually support from what the model asserts is the core of the job. You will work with software engineers and scientists to build benchmark tasks and datasets across viral and bacterial proteins, toxins, RNA elements, and host-pathogen complexes.
Build the Task
- Define what expert reasoning looks like for a structural question and write gold-standard answers with defensible rationale
- Design realistic scenarios across the structural workflow: data collection and processing, map interpretation, model building, validation, assembly and state assignment, comparison, functional interpretation, protein-protein interaction analysis, molecular dynamics simulation, and metadynamics
- Curate datasets across modalities: X-ray structure factors and maps, cryo-EM maps and half-maps, atomic models, AlphaFold predictions, molecular dynamics and metadynamics simulations, NMR shifts and ensembles, SAXS curves, HDX-MS, cross-linking, and SHAPE/DMS reactivity
- Build analysis pipelines in Python (gemmi, Biopython/biotite, numpy) that compute verifiable ground truth directly from the files the agent receives
- Partner with software engineers to turn your judgment into new workflows
Set the Hard Tasks
- Test whether agents distinguish real structure from artifact or over-interpretation: unsupported ligands, register shifts, mirrored maps, crystal contacts read as interfaces, misdocked domains, and B-factors or occupancies treated as evidence
- Adjudicate ambiguous cases: decide what the map or measurement supports, where reasonable experts would disagree, and which additional data would settle the call
- Test whether agents choose the right representation and comparison for the question: biological assembly versus asymmetric unit, the right reference state, and which domain to superpose on
Judge the Agent
- Interrogate agent reasoning step by step and challenge conclusions that would not survive peer review or a validation report
- Pinpoint where agents go wrong: why a step fails, what an expert would do instead, and what evidence would change the call
- Score agent confidence, rewarding explicit uncertainty over overconfident calls on ambiguous structures
Requirements
- PhD in structural biology, biophysics, biochemistry, or a related field
- Hands-on structural experience building and validating models
- Deep expertise in at least one area (X-ray crystallography, cryo-EM, NMR/solution methods, or computational structural biology) and working fluency across the others
- Coding for structural data: compute contacts, superpositions, symmetry mates, map values and geometry from files without a GUI
- AI-ready from day one: daily use of Claude, Cursor, or similar agents
- Confidence to overrule an AI agent when the evidence demands it, and openness to update when shown wrong
- Ability to calibrate confidence and communicate uncertainty
- Bonus: structural virology or antibody-antigen structural work
- Bonus: RNA structural biology, protein-RNA interaction analysis
Benefits
- Catered lunch and dinner daily
- Monthly parking
- Company outings
- Team off-sites
- Run club
- Cycling
- Boxing
- Ice skating
- Soccer
Company overview
LatchBio is a San Francisco-based cloud platform that provides comprehensive data infrastructure and analysis tools for biotechnology research. Founded in 2021, the company offers a unified data platform designed to connect bench scientists, computational scientists, IT teams, and R&D leadership into a cohesive framework. Their platform enables scientists to store, transform, and visualize bioinformatics data, making data analysis faster, more accessible, and cost-effective. LatchBio's solution includes workflow orchestration, scalable cloud computing resources, data storage, and interactive visualization tools specifically tailored for biological data analysis. The company serves hundreds of biotech companies and research organizations, helping them accelerate R&D milestones by eliminating infrastructure complexities. With backing from prominent investors including Lux Capital and Coatue, LatchBio has raised significant funding to develop its platform that supports various omics technologies including NGS, spatial transcriptomics, and single-cell analysis.