🛠 Tech Stack

💼 About This Role

You'll own the hardest model quality problems across our clinical AI products, including foundational coding models and adaptive scribing. You'll drive end-to-end improvements that compound over time, pushing the frontier on latency, accuracy, and cost. This role offers the chance to shape population-level clinical reasoning in real-world healthcare settings.

🎯 What You'll Do

• Own foundational model research on clinical AI problems
• Build compounding learning loops from real-world signals
• Improve Chart Chat quality through smarter retrieval and reasoning
• Optimize latency, accuracy, and cost with appropriate techniques

📋 Requirements

• 5+ years of ML engineering or applied research experience
• Deep expertise in reinforcement learning and deep learning
• Experience shipping model improvements in production
• Strong Python fundamentals and PyTorch experience

✨ Nice to Have

• Publications at top-tier venues (NeurIPS, ICML, ACL, EMNLP)
• Experience with clinical data (EHR, FHIR, medical coding ontologies)
• Open-source contributions to ML libraries or frameworks

🚩 Heads Up

• Role combines research and engineering with high expectations across multiple domains
• No explicit salary range provided, estimated based on title and market