Post details
What does agentic AI development look like once you take it past the demo? Alex Patterson, a Senior Solutions Architect at Sanity and the founder of CodingCat.dev, joins us to talk about giving AI agents verified knowledge instead of guesses, why compiled knowledge bases beat naive RAG, and what changes when an MCP server goes to production, from authentication to agents acting on behalf of other agents. He also shares software engineering career advice for programming through the AI boom from someone who graduated during the dot-com bubble, and why he thinks most AI work could run on local models within a couple of years.Linkshttps://codingcat.dev/post/sanity-knowledge-baseshttps://codingcat.dev/post/the-five-layers-of-agentic-context-why-canon-only-turns-when-someone-signs-offhttps://codingcat.dev/course/mcp-in-productionhttps://codingcat.dev/sponsorshipshttps://codingcat.devhttps://alexpatterson.dev/HostsOvercommitted: https://overcommitted.devBethany Janos: https://www.trustyduck.devBrittany Ellich: https://brittanyellich.com
















