Agentic AI and the Future of AI
summary
Claims that 2025 is the year of agentic AI and demonstrates a working car‑repair and travel agent built in Python using LangChain, LangGraph and LangSmith, with MCP for data plugs, A2A for agent comms, Oracle APEX UI and local models (Olma, Gwen 38B). The demo diagnoses a no‑start car, recommends parts, orders 20 oil filters and schedules mechanic Nora — estimating 5 hours / $520 and asking permission before booking when configured to. Notes resource, privacy and benchmarking limits.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 3:55AI agents replacing jobs
- 5:01What is agentic AI
- 6:24Workflows vs agentic AI
- 9:16Why agentic AI now
- 10:40When to use AI agents
- 13:11Trip booking example
- 15:24Car repair demo
- 22:43Building agentic AI
- 25:10Model Context Protocol
- 26:56Using AI agents in practice
- 28:01Types of agentic AI
- 30:09Developers myth debunked
- 31:54AI limitations
- 32:58Benefits and challenges of AI automation
- 34:03AI agent deployment challenges
- 35:50Future opportunities
- 37:17AI predictions 2040
- 40:49AGI risks and ethics
- 41:31Conclusion and Q&A
related talks
01h 01mKeynote Panel: Future of AI agents
29mAgentic AI: What It Is and Why It Matters
31mBuilding and Orchestrating Production-ready Agentic AI Systems
43mArchitecting the Future: Modern Solution Design with Agentic AI
01h 17mThe year of the AI Agent: Beyond Automation
40mAgentic AI: The Next Evolution of Automation
19mAgentic AI — From Hype to Reality
31mBuilding and Orchestrating Real Production-ready Agentic AI Systems