The World of Vertical AI Agents
summary
Claims most agent demos are LLM wrappers with little production value and pushes for narrow, domain-focused systems. Shows Manis AI on Claude as a quick wrapper example, contrasts it with a built vertical AI agent for data analysts that uses RAG plus SQL routing, guardrails and benchmarking. Reports measured gains: 80% faster data analysis, 20% lower ops cost and 30–40% reduction in inventory waste, and stresses benchmarks determine if it actually works.
AI-written summary · about
chapters (9)
- 0:00Vertical AI agents intro
- 2:07What AI agents are
- 5:19Agents as innovations
- 9:52Vibe coding risks
- 13:03Why vertical agents win
- 17:59When to use vertical agents
- 21:55Measuring agent success
- 23:36Future of AI agent ecosystems
- 25:27Closing remarks
related talks
46mFullstack AI: The Building Blocks for Production Ready AI-Powered Software
02h 08mIntroduction to AI Agents
27mHow to Build AI Agents and Keep Your Sanity
29mBuilding Trust in AI Agents
01h 27mBuilding Vertical Agent with AG2 AgentOS
01h 07mNew Methods of Working with AI.
46mAI Agents for the Rest of Us
07h 25mAI Agents Conference 2026 - Day 2