Your learning library

RT ships real software with coding agents every day (a workspace full of Claude Code skills, client builds, automations) but is light on the engineering theory those skills encode (TDD, domain modeling, deep-module design). The goal is t…

Where we are: Mission established: adopt Matt's eng discipline, practice-first

RT commits daily through Claude Code and Zed but copies git commands without

Where we are: Four-bucket model + add/commit loop established

RT is getting pulled into rooms with Phuket officials (he sat in on a waste-management pitch to a senior official alongside two businessmen). The businessmen could name every department, use the insider slang, and read the hierarchy in t…

Knowledge bases / memory systems for AI agents — including gbrain, Graphiti, and the RAG / knowledge-graph space underneath them.

Where we are: 0003 — Full course built (Lessons 3-11)

Every AI vendor RT builds on holds the client's data on a short clock: VAPI deletes call

Where we are: Starting floor: RT already built pull-and-bank, so the gap is provider-push and S3 as a protocol

RT wants to sell AI agents to enterprise clients whose knowledge lives in big document piles (the live example: an insurance company, currently at talking stage). The DuckyCutz-style live-state agent doesn't fit those clients; a RAG docu…

Where we are: 0002 — Full course built in one session

Who: RT, founder of Rejig Labs. Non-technical on code; thinks in plain English + analogies (memory `feedback_layman_explanations`). Runs a repo full of `.mjs` scripts and `.json` files every day but can't yet read them.

Where we are: 0001 — Mission pivot: voice-eval harness → code literacy