Fishbowl Discussion on APIs and AI
summary
Fishbowl catalogs practical shifts: teams move from REST and microservices to making APIs agent experience ready with machine‑readable specs, AI gateways and policies, and an agentic mesh concept to route events to agents; experiments show surprising scale problems—196 of 200 MCP servers fail compliance and adoption of agentic workflows is around 1%—so groups prototype RAG+LangChain, Gemini, Copilot and mermaid integrations, add observability and “lane assist” guardrails, and debate whether to ship imperfect building blocks or fix technical debt first.
AI-written summary · about
chapters (20)
- 0:00Fishbowl format
- 1:52Legacy systems to API agility
- 3:40Agent experience and AI gateways
- 5:05Event-driven API modernization
- 7:10Domain boundaries and agent access
- 9:19Data exposure for AI agents
- 11:27Data for reliable AI outcomes
- 13:12APIs for agentic AI workflows
- 15:19Agentic AI definition and goals
- 18:55API maturity before AI adoption
- 21:21Policies for agent-API interactions
- 24:35Technical debt vs AI urgency
- 27:44Legacy systems as business enablers
- 29:51Technical debt as strategic leverage
- 31:16Legacy apps with domain-driven design
- 33:04AI adoption communities
- 35:34Event-driven architecture importance
- 39:32AI tooling in engineering workflows
- 42:49Accelerating AI adoption
- 44:35Engineers paired with AI tools
related talks
35mUnleash Your Inner API Architect: Build Agents That Deliver
21mUsing the Power of APIs and Domain Driven Design to Embrace AI [...]
23mDesigning APIs Both Humans and AI Love
21mBingo! Making Sense of AI-Ready APIs in a Buzzword World
26mIf your APIs aren’t agent‑ready, your AI agent is just chat: [...]
21mThe Future of API Management: Autonomous Agents in Action
27mAI-Enhanced API Documentation [...]
43mDesigning REST APIs for the age of AI agents