AI-Driven HATEOAS
summary
A demo builds a cat API on Node that turns static HATEOAS links into dynamic, context-aware actions by calling OpenAI to generate the links at runtime; a prompt enforces rules like “if cuteness score < 8 include feature” and “if not adopted include adoption link,” and the prototype returns adopt or sponsor endpoints correctly in the demo. The claim is that AI-driven HATEOAS makes hypermedia adapt to user behavior, addressing why ~80% of APIs remain REST‑like, and the OpenAI-generated links approach works in the demo.
AI-written summary · about
chapters (10)
- 0:00Introduction to AI-driven HATEOAS
- 1:47Hypermedia as engine of state
- 3:33REST constraints overview
- 6:24Why hypermedia matters
- 9:35Hypermedia missing in APIs
- 16:04Building basic hypermedia API
- 20:43Dynamic hypermedia with AI
- 23:56OpenAI for AI-driven links
- 28:58Summary and future
- 30:04Q&A
related talks
01h 03mMarching Happily Together with HATEOAS!
41mWhy 99% of Companies Building REST APIs Lie?
46mGet HATEOAS and Hypermedia right with Spring
21mBuilding Flexible APIs with Spring HATEOAS
47mHypermedia IRL
46mREST next level: writing business-oriented web APIs
32mAccessibility powered by AI
41mAI without spaghetti: Clean architecture in the age of AI