AI, SQL, and GraphQL Walk into a Fertility Clinic
summary
Compares three concrete architectures—LLM+SQL, LLM+GraphQL, LLM+RestAPI—with runnable Python chains that generate queries, execute them, and validate results; shows prompt structure, attached GraphQL schema, and how resolvers enforce authorization and PII handling; reports why the team chooses GraphQL (exact fetching, secure resolvers) while admitting added development time and cost and listing remaining unsolved edge cases; demonstrates the built feature end to end and how answers are validated against returned database rows.
AI-written summary · about
chapters (13)
- 0:00Fertility clinic challenge
- 1:24Medical data queries hard
- 2:51AI tool for doctor questions
- 6:06Why large language models seemed like a solution
- 8:34SQL security dilemma
- 11:48REST API limitations
- 15:06GraphQL as answer
- 17:12GraphQL solves authorization
- 20:24GraphQL trade-offs
- 21:04AI-GraphQL workflow
- 23:53Validating GraphQL queries
- 25:19AI answers for doctors
- 26:22Key lessons
related talks
29mAI, SQL, and GraphQL Walk into a Fertility Clinic… LLM-based Medical feature development
31mWhy AI needs GraphQL
30mLLM Query Benchmarks: Cypher vs SQL
15mTraining and Using Generative AI Models with GraphQL
18mInteracting with LLMs via GraphQL
50mAI: A Tool for Software Architecture?
01h 59mLLM applications with LangChain
40mTechniques to avoid common LLM pitfalls