Sheet 07 · Engineering blog
Notes from building.
Learnings, tutorials, and the occasional teardown, from the engineering behind our client work. Written by the people doing it.
Jul 2, 2026Two lines to see what your LLM is actually doingBefore we optimize anything in an AI build, we make it observable. Here is the one-minute version with peekr.observabilityllmpeekrJun 24, 2026Giving an AI agent memory that survives the sessionStateless agents forget everything the moment a chat ends. Here is how we give them real, queryable memory with extremis over MCP.agentsmemorymcpextremis