Gamow Labs builds frontier AI systems to interpret the human genome.

We run experiments in our robotic wet lab in Boulder, Colorado.

Research

All research
  1. LabRobotBench V0.2: Measuring Simulation Fidelity for Robot Learning in a Production Lab Workcell

    Grounded in robot logs and real camera footage, we test how faithfully our simulator mirrors a production lab workcell, and show that its realism can be hill-climbed.

  2. ScreenBench: Do AI agents check their data?

    We introduce ScreenBench, a benchmark of AI agents’ performance on compound-screening analysis tasks that contain hidden errors.

  3. Gamow Labs’ Track 2 submission to the Rare Disease, Real Kid: The MVA Hackathon 2026

    For our Track 2 submission, we asked: could an existing drug be used to increase expression of BUB1B and thus the abundance of BUBR1 protein?

  4. LabRobotBench V0.1

    LabRobotBench evaluates how frontier AI agents operate robotic laboratories and recover interrupted wet-lab experiments.

First clinical study: outperformed clinical labs.

In a blinded analysis of 66 alveolar capillary dysplasia cases called non-diagnostic by a number of clinical labs, Gamow Labs successfully identified a molecular explanation for all known cases and even solved two that had evaded human resolution for years (manuscript in review).

66 non-diagnostic cases re-analyzed in a blinded study
100% of known cases given a molecular explanation
2 cases solved that had eluded experts for years

We push frontier models to their limits in genomics and experimental biology.