Free 95-minute deep dive: a Google engineer on building reliable AI agents
goyalshaliniuk · x · 2026-11-14
A free 95-minute deep dive with a Google engineer covers the engineering behind modern AI agent systems, aimed at developers moving beyond basic LLM applications.
Topics covered
- Architecture behind agent harnesses
- Managing context, memory, and tool access
- Connecting agents through MCP
- Running agents in secure sandbox environments
- Handling prompt injection risks
- Measuring and evaluating agent performance
- Building self-improving agents, with a live demo of an agent improving itself
- Optimizing with Google ADK
- Graph-based complex workflows
More from coding & agent
- Free metroidvania level generator: type a seed, get a tiled world exportable as JSON — majidmanzarpour · 2026-11-14
- Replit Agent can now buy for your business, with Visa and Link powering payments — amasad · 2026-11-14
- levelsio's Hotelist adds MCP connector so AI agents can book hotels — alvelda · 2026-11-14
- AWS proposes MixUA: multi-granularity memory scoping for multi-user multi-agent systems — AWS · 2026-11-14
- Apollo Labs launches Tax Law Search MCP with hash-verified citation workflow — Elgon2003 · 2026-11-14
- Developer details why orchestrated coding agents stall mid-backlog and won't chain tasks — W1141175 · 2026-11-14