Building #APIs in the #AI Era
summary
Shows how to build APIs with local and remote LLMs using concrete tools: Podman AI Lab, Olama, Continue, and Spectral. Demonstrates generating an OpenAPI 3.0 spec, iterating it with a local model, then linting (initially 3 errors, an auto-fix run increases to 7), scaffolding Quarkus endpoints from the spec, and running Maven tests. Adds a thin MCP server + client setup (tools exposed to LLMs via SSC/langchain-for-java) so models can call runtime APIs for fresh data rather than hallucinating.
AI-written summary · about
chapters (10)
- 0:00Introduction and AI APIs
- 3:17Using AI for API development
- 6:27Running models locally
- 19:23Open source model considerations
- 22:09API development lifecycle
- 35:34Generating API specs with AI
- 43:45Building APIs from specs
- 45:55Agentic AI and tool integration
- 47:20Model Context Protocol overview
- 58:23Summary and Q&A
related talks
29mPodman AI Lab: For developers to build AI Applications with LLMs running locally 
48mAgentic AI with Quarkus, LangChain4j and vLLM
24mHow Generative AI Apps Use APIs
26mAPIs in the world of Generative AI
28mAPIs and GenAI : Your Perfect Combo with LLM-Powered Frameworks.
57mPure Java Enterprise AI/LLM Integration (EAI 2.0)
40mAPI Lifecycle: Smarter Design and Governance With Open Tooling
41mAI without spaghetti: Clean architecture in the age of AI