About the Role:
- You decide the correct answer for each case and build the studies that test it.
- You review AI agents handling medical images, identifying missed and false findings.
- You work across MRI, CT, ultrasound, and radiography to design benchmarks.
Requirements:
- Professional experience with thousands of medical imaging studies, deep expertise in at least one modality.
- Comfortable with Python and daily use of AI agents like Claude Code.
- Ability to judge technical correctness and clinical unsoundness.
Culture @ Latch:
- We practice radical transparency, always us against the problem, never each other.
- We use AI constantly and reward original expert judgment.
- Outside work, we have company outings, team off-sites, and a community with various activities.
Latch
Latch builds rigorous scientific benchmarks for AI agents across biology. We are a small, transparent team that values collaboration and continuous improvement.