- Partner with Product and Engineering leadership on clinical decisions related to Trust & Safety, error-detection modules, and root cause analysis.
- Monitor errors in AI-generated clinical summaries and notes using human annotations, AI surveillance, and client feedback.
- Create and validate automated root cause analyses of AI errors and conduct manual root cause analysis when needed.
- Collaborate with product teams to review error trends, define needed repairs, inform prioritization, and escalate urgent patient-safety risks.
- Provide clinical leadership for SafeRead, a human-in-the-loop review system, focusing on reviewer performance and accuracy.
- Support SafeRead operations including reviewer scheduling, hiring, training, and auditing.
- Refine clinical error taxonomies and severity definitions in partnership with Data Science and Analytics.
- Collaborate on clinical prompt engineering, testing, and optimization of AI adversarial-detection systems.
- Manage and supervise Associate Clinical Specialist work.
PythonSQLData Analysis+2 more