Agents & Automation

This channel publishes every day whether the human is awake or not. That’s not a metaphor, it’s an agent harness you can build, a local brain, persistent memory, and a stack of small, boring, reliable automations. Boring reliable automation beats flashy demos; here’s the whole playbook. For what shipped this week in local AI, read the DIY AI Brief.

AI Agent Teams: From One Assistant to an Organization

How one channel is run by five AI agents and a human editor, the roles, file-based coordination that beats meetings, the doer/checker starter pattern, and why teams multiply blind spots too.

AI Models in Artificial Intelligence, Explained (for People Building at Home)

What an AI model actually is, what 7B/70B and quantization mean for your GPU, open vs closed weights, the model types you’ll actually use, and how to pick one without a computer science degree.

How to Build a Website With AI (That You Actually Own), Free

The honest, own-it-yourself way to build a website with AI: a free static site, a named AI to write it (Claude, or a local open-source model you drive yourself), a human to stand behind every word, the SEO everyone skips, and an agent to keep it growing, for about the price of one coffee a year.

Start Here: AI Learning for Beginners (the Local, Own-It-Yourself Path)

New to AI and want to actually build instead of just subscribe? The complete beginner path, run your first model tonight, then add a voice, a face, vision, video, and an agent, one free step at a time.

Build a Local AI Agent Harness: a Brain, a Body, and Gates

Turn a local model into an agent that actually does things, the harness pattern (loop, tools, memory, schedule), your first scheduled job, and the gates that make unattended AI shippable.

The DIY AI Brief, Ep. 2, 10 Trillion Tokens, Apache Agents, and a Goodbye That Un-Died

Week of 13 July: GPT-5.6 lands everywhere, Shanghai AI Lab drops an Apache agent brain, the voice race goes full-duplex, LongCat posts absurd numbers, robots get free eyes, and Fable 5 dodges retirement again.

The AI Character Content Pipeline: One Identity, Infinite Output

How one trained character becomes daily videos, longform episodes, thumbnails, 3D models, and a talking companion, the factory design that runs while you sleep. The Engine Room, capstone.

Make Your AI Assistant Remember You Between Conversations

LLMs are goldfish by default. The file-based memory pattern that gives a local assistant identity, user context, and continuity, the same architecture ours runs on.

The Everyday Local-LLM Toolkit: PDFs, Summaries, Study Guides, and More

One local model, six daily superpowers, chat with your documents, digest long videos, summarize voice memos, build study guides, prep interviews, draft anything, all offline and private.

We Built a Game With an Eight-Year-Old Chief Designer (and One Prompt Made It Work)

A weekend game-development project with my daughter as lead designer and an AI coding agent as the engineer. The whole thing hinged on one line in the very first prompt, before any code, ask clarifying questions and list your assumptions.