I Built an AI Assistant That Doesn't Need the Internet

You are helping me turn one specific YouTube video into real, durable learning.

Source video:
- Title: I Built an AI Assistant That Doesn't Need the Internet
- URL: https://www.youtube.com/watch?v=RTtVIW36bd4
- Topic: Creative Automation
- My current learning frame: Spec your own minimal offline AI device on paper — pick a single-board computer, input method, and battery under a $150 budget, state three design constraints, and predict tokens-per-second and runtime before you buy anything.
- Why this matters: New playlist item from Built By West; queued for transcript-backed review, topic mapping, and a practical learning artifact.

Transcript anchors from this exact video:
- 0:18 / Evidence 1: ">> This is what I'm calling the Think Slab. It's a fully portable battery-powered computer designed to run AI models. It's been my passion project for a while, so a lot of work has gone into creating it."
- 2:14 / Evidence 2: "Earth has been condensed into an AI assistant that fits in the palm of your hand. The AI could help me choose. I poured every thought, every detail, every variable into my prompt, desperate for at least a..."
- 4:49 / Evidence 3: "possible. I want to keep it smaller than an iPad, but not so small to the point where I can't use a touchscreen keyboard. Next, it needs to be simple. According to my favorite industrial designer, Dieter Rams,..."
- 6:35 / Evidence 4: "by a Chinese company called PiSugar, was just what I needed. Even though it only carried a quarter of the power, it wasn't only dirt cheap, but it was also so compact that it could shrink my initial..."
- 8:49 / Evidence 5: "2B, which I found is the best balance of accuracy and speed for this setup. >> I should mention that the performance will not be considered good by today's standards. This is a $50 computer with no dedicated..."
- 10:32 / Evidence 6: "you guys need to know about me is that I procrastinate sometimes, a lot. So, I designed the Think Slab right here back in October of 2025. This was before I knew how to use GPIO and other..."
- 12:09 / Evidence 7: "not only increase longevity, but up the performance a lot. I've already looked into using those pre-made heat sinks and coolers with fans attached to them, but the way they were designed would require a larger case. Lastly,..."

Video-aware target:
- Prompt lane: Creative automation
- Mechanism to extract: Extract the creative production loop, especially where the human keeps taste, selection, and final judgment.
- Artifact to produce: A creative production board with source inputs, prompt recipe, selection criteria, edit pass, and taste-review checkpoints.
- Artifact must include: brief; source inputs; generation recipe; selection criteria; edit/review checkpoint

Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, transcript support, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable mechanism from the video: Extract the creative production loop, especially where the human keeps taste, selection, and final judgment. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A creative production board with source inputs, prompt recipe, selection criteria, edit pass, and taste-review checkpoints.
5. Include:
   - a plain-English definition of the core idea
   - a diagram or structured model using this sequence: Brief -> Source material -> Generation -> Selection -> Edit -> Taste review -> Reusable recipe
   - answers to these source questions: What asset is being produced? | What inputs and tools drive it? | Where does human taste intervene?
   - 3 concrete examples that apply the video idea to real agentic work, such as Claude-generated video campaign; image-to-site workflow; voice or video editing loop
   - 2 failure modes the video helps prevent, chosen from the transcript evidence and these likely risks: mistaking novelty for quality; no source/brief discipline; shipping generated media without taste review
   - a checklist for the next real workflow, focused on: brief, inputs, generation, selection, critique
   - one practical exercise with a clear done signal: Build one reusable creative recipe and define what would make the result rejectable.
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.

Quality bar:
- Make this specific to "I Built an AI Assistant That Doesn't Need the Internet", not a generic Creative Automation essay.
- Anchor each creative step to transcript evidence about inputs, model/tool choices, iteration, editing, or critique.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- Avoid these generic drifts: generic content advice; tool hype; creative output without selection criteria.
- If evidence is weak or missing, stop and say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.