Apr 18, 2026 · 59m · startup-ideas
Claude Design: Full Walkthrough. I'm blown away.
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this unfiltered live walkthrough, Greg Isenberg rigorously tests Anthropic's Claude Design across mobile UI prototyping, pitch deck generation, and video creation. While demonstrating remarkable strengths in wireframing, strategic slide decks, and canvas editing, the test also exposes notable limitations in parallel task processing and video rendering.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Greg holds 100% of the talking time here. How this is scored →
Greg as informed peer 5.0 Guest teaching 0.0 Guest disagreement 0.0 Greg pushing back 0.0
speaking balance: gold is Greg, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 57:30 Direct critique of video rendering quality
In a solo livestream without sparring guests, the sharpest critical push comes from Greg bluntly declaring that the tool's video output 'sucks' and is not ready for prime time.
Hardest push from Greg ▶ 44:50 Host pushes back on missing UI questionnaireGreg firmly criticizes Claude Design's UI for auto-dismissing and skipping custom setup questions rather than letting the user complete them.
Biggest teaching moment ▶ 25:00 Discovering multi-project concurrency limitsGreg learns in real time that running simultaneous generation prompts across separate project tabs freezes the active pipeline.
Greg holds their own ▶ 39:05 Applying social growth expertise to UI copyGreg demonstrates his product expertise by explaining why copy reading 'share this win' vastly outperforms generic social share buttons in user conversion.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Greg as informed peer | Guest teaching | Guest disagreement | Greg pushing back | Why |
|---|---|---|---|---|---|---|
| Livestream Kickoff and Starting the Exploration | 3 | 0 | 0 | 0 | Greg introduces the livestream format and sets expectations for an unedited, real-time product test of Claude Design. There are no actual guests present, aside from a brief greeting exchange with chat. | |
| Ideation and Wireframe Setup for Senior Brains | 6 | 0 | 0 | 0 | Greg demonstrates product management domain expertise by structuring app constraints, advising against mid-fi wireframes, and setting up targeted questions for senior accessibility. | |
| Evaluating Three Wireframe Directions with Live Chat | 6 | 0 | 0 | 0 | Greg reviews three distinct UX directions, comparing them to the deliverables of a design agency while drawing on his background running an agency building AI applications. | |
| High-Fidelity Prompting, Bug Encounter, and Pitch Deck Initiation | 4 | 0 | 0 | 0 | When Claude Design encounters an error during hi-fi generation, Greg troubleshoots the bug live and pivots to generating an investor pitch deck in parallel. | |
| Pitch Deck Configuration and Parallel Processing Limitations | 5 | 0 | 0 | 0 | Greg explores multi-tasking limitations within Claude Design, discovering that concurrent tasks stall the pipeline, while reading debugging suggestions from the chat. | |
| In-Depth Review of the Generated Sequoia Pitch Deck | 7 | 0 | 0 | 0 | Greg dissects the AI-generated Sequoia pitch deck slide by slide, evaluating market sizing, unit economics (CAC, LTV, payback), and strategic GTM messaging with startup expertise. | |
| Reviewing Hi-Fi Wireframes and Testing UI Annotations | 7 | 0 | 0 | 0 | Greg reviews hi-fi screens and uses the annotation tool to request a Facebook share button, noting how framing copy as 'share this win' optimizes conversion based on his social product experience. | |
| Setting Up the Video Ad Prototype and Interface Navigation | 4 | 0 | 0 | 0 | Greg attempts to prompt a video prototype and critiques interface friction when navigation loses context and dismisses questionnaire forms automatically. | |
| Waiting on Video Generation, Health Break, and AI Tool Comparisons | 3 | 0 | 0 | 0 | While waiting for the video build, Greg takes a turmeric health drink break, interacts casually with live chat comments, and previews alternative AI video generation tools. | |
| Reviewing the Video Prototype and Assessing Video Generation Quality | 5 | 0 | 0 | 0 | Greg reviews the final video output and delivers a candid concluding assessment, rating wireframes and decks as top-tier while video capabilities are mediocre. |
Statements from this episode (11)
Isenberg: Mid-fidelity wireframes are bad; use low-fi or high-fi instead
“Mid-fi wireframes are bad. You know, I think that you want to start with low-fi or go high-fi. You know, you can't be half pregnant. You don't want to go mid-fi.”
Isenberg: Senior Brain-Training App Could Generate $5M to $15M ARR
“If you can actually build that, like I could see how you, this could be a five, 10, fifteen million dollar ARR business on top of it, right? You start with Instagram reels and you probably focus on Facebook. Like the seniors are on Facebook.”
Isenberg: Challenging LLMs With Concrete References Improves AI Generation Results
“So I find that that's helpful just to challenge any LLM giving it references.”
Isenberg: AI Video Tutorials Frequently Edit Out Tooling Glitches and Errors
“Because when you go on YouTube, and you see tutorials about You know, this is insane. This product's insane. And I've, you know, I've been guilty of this too. Sometimes you don't, you know, you don't put this in. You don't put the errors in. This gets edited o…”
Isenberg: Claude Design created the best LLM-generated pitch deck he has seen
“So this is probably the best deck I have ever seen created by any LLM period. Period.”
Isenberg: 'Share this win' CTAs convert significantly better than generic share buttons
“I didn't tell it to say share this win, but I know as someone who's built social products that millions of people have used, That if you say share this win versus share this on Facebook, share this win is going to convert a lot more. And what does that mean ta…”
Isenberg: Claude Design losing context when switching folders is a UX miss
“What I do find confusing is every time you like leave a folder and go back in, it kind of like it loses context a little bit. See, I wanted to get back to the questionnaire but when I went back here, there was nothing, right? But anyways, I gave it the, so tha…”
Isenberg: Answering AI design questionnaires forces PMs to clarify their thinking
“One of the things I love most about Claw Design, funny enough, as a first reaction, is doing the questionnaires. Because in doing the questionnaires, I, as like the product manager, can get so much more clear as to what exactly I'm building. Of course, I grabb…”
Isenberg: Claude Design output delivers thousands of dollars in value
“Like the value right now has been thousands of dollars. Period. The value has been thousands of dollars. Did you see that deck? That deck was crazy. The directions, the wireframe directions were on point, and the visual style was really good. It nailed 90% of …”