Unlocking Agentic AI: [...]
summary
Claims that 60% of CEOs mandate AI agents while 65% admit they’re unready and that 95% of AI projects fail drive the talk’s main push: build a trusted platform with three concrete layers—agent runtime (LLMs + RAG + vector DBs), integration via an MCP gateway to legacy APIs and mainframes, and governance (LLM, MCP, A2A gateways). Practical rules: prototype with mocked endpoints, limit tools to ~5–15, and phase in API/data/AI governance before scaling to production.
AI-written summary · about
chapters (18)
- 0:00Agentic AI introduction
- 1:33Agentic enterprise vision
- 3:17AI timeline and integration
- 5:08AI prototypes vs adoption
- 6:34Trusted platform for AI agents
- 7:15Critical AI platform capabilities
- 8:38AI agents vs LLMs and RAG
- 11:42AI agent architecture
- 13:29Enterprise AI integration challenges
- 15:16Connecting AI agents via MCP
- 16:42Governance for AI agents
- 18:11Enterprise architecture for AI
- 20:20Gateways in AI architecture
- 21:48Managing API and MCP sprawl
- 24:03API and MCP governance
- 26:10Summary and next steps
- 26:54Q&A scaling AI proof to prod
- 29:02Q&A merging governance frameworks
related talks
29mBuilding Trust in AI Agents
21mOvercoming Data Barriers for Successful AI Projects
28mAgentic Enterprise - What your CEO must know about AI
24mAI Governance for developers: A practical framework for deploying autonomous coding agents safely…
30mAgentic AI in Action: From Framework to Deployment
28mBuilding an AI-Ready Workforce
20mImplementing Agentic AI Systems That Redefine Industries
45mDeploy First, Secure Never: AI Between Productivity and Security Chaos