United StatesHealthcare and life sciences2025 – ongoingChildren's National Hospital
Spotlight AI assistant for staff questions on policies and procedures
Children's National replaced 30 to 45 minutes of daily document review with an AI policy chatbot
Problem
Staff at the Washington, D.C. pediatric hospital had to read through policy and procedure documents to find answers for onboarding and accreditation. The review took 30 to 45 minutes a day.
Approach
The hospital built Spotlight AI, a retrieval-augmented generation agent, on the Databricks platform. It answers staff questions about policies and procedures and links to the source documents. The hospital tracks and evaluates its models and prompts with MLflow.
Results
Takeaway
A document-search assistant that cites its sources gave staff a faster route to policy answers. The hospital did not say how it measured the time saved.
Categories
- Industry
- Healthcare and life sciences
- Business function
- Internal work and knowledge management
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Databricks describes how Children's National uses a retrieval-augmented generation agent to answer staff questions about policies and procedures. The page says document review that took 30 to 45 minutes a day is now instantaneous. It also mentions forecasting models and a clinical risk tool still in validation, without result figures.
Full details are in the original sources.