ThesisDesign a Fully Animated Website with AI — Claude Code teaches a practical interfaces + open design move: This video shows how the creator built a fully animated portfolio website in about 2 hours by sketching states in Figma, generating warm-toned model photos with Nano Banana/Hexfield, and prompting Claude Code section-by-section to produce a flip-to-ticker intro, project pages, and an info page.
The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.
0:51Sketch as backup
“because I prepared a small free gift to you as a thank you for watching. Before I start prompting in Claude code, I sketched the concept for the website in Figma. And the purpose of these sketches is...”
Figma sketches aren't meant to be replicated pixel-for-pixel; they serve as a visual fallback Claude consults only when a text prompt is ambiguous, since text prompts are prioritized over the reference images. Sketch rough state references for a small layout, then write text prompts that carry the precise details and note which details you deliberately left to the prompt rather than the sketch.
1:54Build animations in states
“these sections. This is the last design that I created for the about me page and I inverted the colors here just to keep things interesting and clearly distinguish it from the project pages. And I went for...”
Advanced transitions are prompted by describing a discrete opening state, a transition, and a resting state, so Claude can interpolate a smooth motion between two clearly defined endpoints (e.g. flip wordmark to spread-out ticker row). Pick one animation idea and rewrite your prompt as three labeled sections (opening state, transition, resting state) before generating any code.
5:31Confirm before fixing
“please make the adjustments." And then it worked perfectly. And this is just one of those small techniques that can really improve your prompting by asking the model if it understands you before it starts making changes. Once...”
When a result is wrong and the cause is unclear, explain what you think went wrong and ask Claude to restate the issue back before requesting changes; its restatement often clarifies the problem better than the original description and leads to a correct fix. On your next broken AI output, withhold the fix instruction and first ask the model to explain the issue in its own words, then approve only once it matches your intent.
01Intent
Start with this video's job: This video shows how the creator built a fully animated portfolio website in about 2 hours by sketching states in Figma, generating warm-toned model photos with Nano Banana/Hexfield, and prompting Claude Code section-by-section to produce a flip-to-ticker intro, project pages, and an info page. Treat "Intent" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:51, where the video says: “because I prepared a small free gift to you as a thank you for watching. Before I start prompting in Claude code, I sketched the concept for the website in Figma. And the purpose of these sketches is...”
02Canvas
Use "Canvas" to locate the part of the interfaces + open design workflow the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 1:54, where the video says: “these sections. This is the last design that I created for the about me page and I inverted the colors here just to keep things interesting and clearly distinguish it from the project pages. And I went for...”
03Artifact
Turn "Artifact" into the reusable artifact for this lesson: A UI critique sheet for judging whether an AI interface improves control. This is where watching becomes something you can inspect and reuse.
04Preview
Use "Preview" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.
05Feedback
Use "Feedback" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.
06Iteration
Use "Iteration" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.
ExampleSource-backed work packet
Convert the video into a scoped task that includes the transcript claim, target workflow, acceptance criteria, and proof. The output should be a ui critique sheet for judging whether an ai interface improves control..
ExampleClaim vs. demo brief
Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the workflow.
ExampleTeach-back module
Transform the lesson into a definition, a mechanism diagram, one misconception, one practice exercise, and a check-for-understanding question.
Do not learn it wrong- Treating the title as the lesson without checking what the transcript actually says.
- Letting the prompt drift into generic advice that could apply to any video in the playlist.
- Copying the tool setup without identifying the operating principle that transfers to your own stack.
- Skipping the artifact, which means the learning never becomes operational or inspectable.