Medical AI Researcher

Tessel conducts rigorous, independent evaluations of medical imaging AI systems, bridging the gap between benchmark performance and clinical reliability. We analyze how models behave in real-world settings — where they generalize, where they fail, and what uncertainties remain — serving as a trusted independent evaluator for both healthcare institutions and AI vendors.

As a Medical AI Researcher, you will work directly with medical imaging companies preparing for FDA 510(k) or De Novo submissions. You'll own customer engagements end-to-end: running meetings, defining evaluation questions, leading investigations, and delivering evidence that materially affects go/no-go decisions. The focus isn't building models — it's understanding and demonstrating their behavior. Where does the model generalize? Where does it break? What trade-offs are being made? What uncertainty remains? The output is defensible behavioral understanding, supported by rigorous evidence — clear enough to inform internal decisions, build customer confidence, and withstand regulatory review. You'll combine practical ML skills with customer-facing judgment and scoping discipline, delivering decision-grade evidence reliably under time pressure. This role requires hands-on expertise in medical imaging workflows and integration, including DICOM/PACS and radiology pipelines, along with strong empirical ML instincts — designing investigations, not just training models. You should be comfortable reasoning about model behavior, failure modes, and uncertainty, and able to work fluently in Python from messy real-world data through to defensible conclusions.

This is an early-stage role with significant autonomy and company-level impact, offering a competitive salary of $150k–$250k+ and meaningful equity (1–3%).