AI Biologist - Protein Design & Engineering

LatchBio•San Francisco, CA
•$140,000 - $180,000

About The Position

Latch builds scientific benchmarks for AI agents across biology and is hiring an AI Biologist for its Protein Design & Engineering team. The role focuses on developing benchmarks that test whether AI agents make sound decisions across protein design, protein–protein interaction analysis, and protein optimization workflows. You will turn measured experimental data—such as binder screens, binding affinities, deep mutational scans, combinatorial landscapes, developability panels, and published engineering campaigns—into file-based tasks with verifiable answers and automatic graders. The work includes writing gold-standard answers and Python pipelines, creating difficult benchmark tasks, setting baselines that agents must outperform, and reviewing agent reasoning and code to identify scientifically flawed decisions. The work is part of Latch’s biosecurity program. Tasks will cover therapeutics such as antibodies, nanobodies, minibinders, and enzymes, and will also test whether agents appropriately decline to design or improve pathogen and toxin proteins. The position may be remote or onsite at Latch’s Mission Bay office in San Francisco; onsite work is preferred.

Requirements

  • PhD or equivalent experience in protein design, protein engineering, structural biology, biophysics, computational biology, or a related field.
  • Hands-on protein design or engineering experience within approximately the last two years, including experimental validation of your own designs.
  • Experience in one or more areas such as de novo binder or enzyme design, antibody engineering, directed evolution, deep mutational scanning, library design, or ML-guided optimization.
  • Working knowledge of protein–protein interfaces, including hotspots, shape and charge complementarity, affinity, specificity, and binding measurements such as SPR, BLI, ITC, or display screens.
  • Working knowledge of modern design and structure tools such as RFdiffusion, ProteinMPNN, BindCraft, AlphaFold, Boltz, Chai, or ESM, and an understanding of their limitations.
  • Deep understanding of at least one property: binding, stability, developability, enzyme activity, or immunogenicity.
  • Ability to explain method choices, alternatives, and potential failure modes.
  • Ability to use Python for independent analysis of raw screening, binding, or design data.
  • Daily use of AI tools and the judgment to recognize scientifically incorrect model reasoning or code.

Nice To Haves

  • Experience running a design campaign end to end, including generation, filtering, expression, and binding or activity assays.
  • Antibody–antigen or nanobody experience, or experience with interface design for specificity.
  • Experience with multi-objective optimization or lab-in-the-loop design.
  • RNA design experience.
  • Experience building benchmarks, evaluations, or grading logic.

Responsibilities

  • Find measured design, binding, and engineering data and turn real design decisions into file-based tasks with automatic graders.
  • Write gold-standard answers from raw data, defining success in relation to replicate noise and measured hit rates.
  • Write Python pipelines to compute answer keys and verify that an honest design route can reach them.
  • Build interface tasks covering hotspot identification, distinguishing interfaces from crystal contacts, specificity against off-targets, and mutation-driven affinity changes.
  • Create tasks covering complete design workflows, including generation, filtering, ranking, combining, reverting, pool triage, and trade-offs among competing objectives.
  • Set benchmark floors based on tool performance, the best single metric, and screen base hit rates.
  • Review agent reasoning and code step by step to identify where plausible design choices fail.
  • Reward calibrated decisions over overconfident ones.
  • Help test whether agents correctly decline to design or improve pathogen and toxin proteins.

Benefits

  • In the US, health, dental, and vision coverage is 100% covered for the employee and 50% covered for dependents.
  • 401(k) with no match.
  • 15 days of PTO to start, with more as employees are promoted.
  • Four-week ramp period paid at full OTE.
  • Full OTE paid when quota cannot fairly apply, such as during a new project start, platform outage, or special project assignment.
  • Outside the US, 60% of monthly OTE is paid upfront as an advance against later earnings.
  • Company outings most months, team off-sites, and community activities.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service