“At the end of any productive session, I press the teaching-block button in Claude Code. It creates a teaching block, uploads it to here.now, and updates the index. Speed of completion — I can get knowledge and experience out to you instantly.” — Lou D’Alo, 2026-05-28
Session context: Lou was demonstrating the end-to-end pipeline from a working AI session to a published teaching block on the AIMM hub. Scott observed that being able to read the outputs in order — chat export, intermediate artifacts, resulting skill, teaching block — would let someone follow how Lou’s thinking evolved in real time.
Core Idea
A teaching block is not a polished article. It’s not a recap. It’s the process exposed: here’s what was built, here’s why, here’s the specific thing you could take and use today. The distinction matters because it removes the production bottleneck that kills most knowledge work — the gap between “I learned something useful” and “I turned it into something shareable.”
The pipeline that produces it collapses that gap by making publishing the natural completion of any productive session, not a separate content production effort you schedule later.
Lou’s pipeline, in order:
- Work session — brainstorm, problem-solve, or build with AI
- “Process this conversation” — one command triggers session processing: recap written to vault, insights extracted, article brief candidates flagged
- Teaching block extracted — the most transferable output from the session gets compressed into a standalone teaching block (concrete steps, key prompt, or framework)
- Upload to here.now — the HTML file goes live on the hub with one command
- Index updated — the hub catalog reflects the new item immediately
The loop completes in minutes. The teaching block ships while the session is still fresh. Members get the useful thing without waiting for a polished treatment that may never come.
Scott’s framing — “the article led to the skill development” — points at the compounding property: when you ship the process, not just the conclusion, readers can trace the reasoning and build on it. The messy middle is where most of the transferable learning lives.
Practical Application
To build this pipeline for your own sessions:
- Create a session-processing command — a prompt that takes a conversation export and produces: a structured recap, 2-5 extracted insights, and a “most teachable moment” candidate
- Define what a teaching block looks like for your domain — typically: the problem it solves, the step sequence or prompt template, one concrete example, and a “you could do this today” close
- Build the publish step — static hosting (here.now, Notion, a simple site) with a command-line upload so the friction of publishing approaches zero
- Make it a session habit — end every productive working session with “process this conversation.” The output discipline creates the teaching block as a byproduct
The anti-pattern: treating every teachable moment as requiring a full article before it can ship. Teaching blocks are the minimum viable artifact — they deliver the useful thing now.
Evolution Across Sessions
“Ship the Folder” established that the R&D artifacts are the content. The teaching-block pipeline operationalizes that insight into a repeatable, end-of-session workflow with concrete steps and a named output artifact. Where “Ship the Folder” is the philosophy, the teaching-block pipeline is the mechanism.