Member of Technical Staff

What you'll do

• Build and scale agent systems for clinical development workflows.
• Own evaluation infrastructure and methods, including performance measurement and failure analysis.
• Use evaluation results to guide agent improvements from development through production.

What they ask for

• Experience building and operating agent systems at scale in rigorous use cases or regulated environments.
• Hands-on ownership of evaluation infrastructure and methods, including task benchmarks, regression testing, and production monitoring.
• Strong software engineering and ML fundamentals, including diagnosing failures across models, tools, retrieval, and workflows.
• A disciplined approach to reliability, evidence, and human review in consequential workflows.

Nice to have

• Experience with model post-training, including supervised fine-tuning, preference optimization, or reinforcement learning.
• Hands-on modeling and experimental design linking model changes to measurable improvements in agent performance.

Tools

agent systems, task benchmarks, regression testing, production monitoring, supervised fine-tuning, preference optimization, reinforcement learning

AI focus

Build and evaluate agents for clinical development workflows, including post-training and modeling to improve performance.