Advancing Drug Search with PostgreSQL and Azure AI
summary
Built a mobile drug‑search app in 5 days that wins the TIDE NATO Hackathon and gets a grant, using PostgreSQL with the azure_ai extension to store vectors and power a RAG assistant. Search layers combine ILIKE/LIKE queries, trigram fuzzy matching via pg_trgm, and vector similarity over a corpus of ~700,000 brand names. Uses PostgREST, ATC categorization, barcode/photo/voice capture and returns dosage/equivalents reliably in field tests.
AI-written summary · about
chapters (7)
- 0:00Intro
- 1:02TIDE NATO Hackathon challenge
- 3:01Advisors & Problem Definition
- 10:26Technical Specifications
- 25:22Multi-layered approach
- 26:41Link to demo
- 27:58Contact info
related talks
46mNATO Hackathon Winner: AI-Powered Drug Search
44mPostgres and OpenAI - a match made in Azure
12mThe mechanics of integrating AI and vector search into Azure Database for PostgreSQL
24mFrom Queries to Agents: The Next Era of Data Retrieval on PostgreSQL
48mThe power of AI and Copilot for Azure Databases
42mUsing Postgres to locate the best coffee near you!
50mBuilding AI ready applications
47mExtendable by Design: Building Generative AI Apps with Postgres and Vector Storage and Azure AI