Becker’s Healthcare Podcast Scaling AI Innovation at Seattle Children's with Dr. Zafar Chaudry
Feb 24, 2026
Dr. Zafar Chaudry, a physician and Chief AI and Information Officer with 40 years in healthcare, describes Seattle Children’s AI journey. He talks about migrating data to the cloud and consolidating systems. He outlines an agentic Pathway Assistant for clinicians. He covers translation and ambient AI, cybersecurity, cost and app rationalization, and patient-centered app features.
AI Snips
Chapters
Transcript
Episode notes
Data Platform First Enables Reliable AI
- Seattle Children's rebuilt its data platform on Google Cloud and fed a data lake from 171 systems to enable accurate AI models.
- They use Gemini and Vertex AI to train models and prioritize getting the right data into one place before modeling.
Involve Clinicians Continuously When Building AI
- Do involve end users early and continuously when building AI; Seattle Children's formed an AI review board and mandatory AI training.
- Validate use cases with clinicians and run continuous PDSA cycles to sustain accuracy and adoption.
Turning Thousands Of Pathway Pages Into An AI Guide
- Seattle Children's converted 6–7,000 pages of validated clinical pathway PDFs into an agentic AI called Pathway Assistant.
- Fifty physicians contributed ~100 hours each to build, validate, and test the agent that guides diagnosis, tests, and treatment for pediatric cases.