Got laid off from Block last Friday. Instead of pausing, I’m pivoting. 🚀
I’m officially starting my journey to build in public, starting with an app I've been wanting to make for my three kids.
Introducing Story Fabric 🧵✨ A storytelling engine that turns wild bedtime improvisations into vibrant, illustrated scenes instantly. Just voice input and imagination—no typing required.
I would love to get your thoughts on the concept and the UI design attached!
To the other 4,000+ impacted: let's turn this into a launchpad. What are you building? Drop it below. 👇
#blockmafia #buildinpublic
Rethinking how we build for the agentic era. Stop writing agent skills like human docs—it wastes tokens and invites hallucinations.
Here's a distilled set of core best practices (<5 min read) to help us write better skills.
Give it a read: https://t.co/6ep4VXg2zJ
Day 4 of building my AI kids storybook app:
Today I moved from “demo-ish” to real product infrastructure:
deployed the backend live on DigitalOcean
fixed story list persistence/rehydration behavior in the app
fixed page generation/render resume issues
removed fake/mock UI from core screens
added a real benchmark harness for fast-sdxl, flux/schnell, flux-subject, and Gemini image lanes
Next goal: get reader-mode images under 3s, and reserve HD generation for print/refine only.
#BuildInPublic #blockmafia #ios #mobiledeveloper
Day 3 #buildinpublic update: The iPhone app is WORKING! 📱✨
The end-to-end flow is alive: Speak a prompt -> Gemini generates story text -> App builds the book pages.
But we hit a wall. 🚧
Currently, generating the images takes ~15 seconds per page. That's too slow for an impatient 6-year-old. The magic dies if you have to wait.
The Pivot: Instead of generating 6 images sequentially, I'm refactoring the backend today to use a "Contact Sheet" technique.
I'm switching to the Flux model on https://t.co/Du1OGtqoZl to generate one GIANT grid image containing all story scenes simultaneously, then slicing it up on the server.
Goal: Drop generation time from 15s down to sub-5s for the entire story.
Here are screenshots of the actual V1 app running on device. The comic book vibe is coming together! 👇
#blockmafia #swiftui #aiengieering #iosdev
Hey everyone, Day 2 update! I spent today locking in the architecture for Story Fabric and putting together a quick product video to show the flow
The biggest challenge with AI storytelling apps? The waiting room. Kids (and parents) don't want to stare at a spinner.
My latency strategy for Story Fabric: ⚡️ Gemini 2.5 Flash generates the full text story instantly. 🔀 Next.js + Inngest queues parallel image jobs (Imagen 4.0), prioritizing Pages 1 & 2. 📡 Supabase Realtime progressively reveals the artwork in the SwiftUI app as it renders.
The goal? First illustrated page on screen in under 6 seconds.
Checkout the V1 flow below. What do you think of the Neo-Brutalist UI? 👇
#blockmafia #buildinpublic #swiftui
Got laid off from Block last Friday. Instead of pausing, I’m pivoting. 🚀
I’m officially starting my journey to build in public, starting with an app I've been wanting to make for my three kids.
Introducing Story Fabric 🧵✨ A storytelling engine that turns wild bedtime improvisations into vibrant, illustrated scenes instantly. Just voice input and imagination—no typing required.
I would love to get your thoughts on the concept and the UI design attached!
To the other 4,000+ impacted: let's turn this into a launchpad. What are you building? Drop it below. 👇
#blockmafia #buildinpublic
I attended a vibe coding hackathon recently and used the chance to build a web app (with auth, payments, deploy, etc.). I tinker but I am not a web dev by background, so besides the app, I was very interested in what it's like to vibe code a full web app today. As such, I wrote none of the code directly (Cursor+Claude/o3 did) and I don't really know how the app works, in the conventional sense that I'm used to as an engineer.
The app is called MenuGen, and it is live on https://t.co/bQonQT88t0. Basically I'm often confused about what all the things on a restaurant menu are - e.g. Pâté, Tagine, Cavatappi or Sweetbread (hint it's... not sweet). Enter MenuGen: you take a picture of a menu and it generates images for all the menu items and presents them in a nice list. I find it super useful to get a quick visual sense of the menu.
But the more interesting part for me I thought was the exploration of vibe coding around how easy/hard it is to build and deploy a full web app today if you are not a web developer. So I wrote up the full blog post on my experience here, including some takeaways:
https://t.co/2kkQh0ElgB
Copy pasting just the TLDR:
"Vibe coding menugen was exhilarating and fun escapade as a local demo, but a bit of a painful slog as a deployed, real app. Building a modern app is a bit like assembling IKEA future. There are all these services, docs, API keys, configurations, dev/prod deployments, team and security features, rate limits, pricing tiers... Meanwhile the LLMs have slightly outdated knowledge of everything, they make subtle but critical design mistakes when you watch them closely, and sometimes they hallucinate or gaslight you about solutions. But the most interesting part to me was that I didn't even spend all that much work in the code editor itself. I spent most of it in the browser, moving between tabs and settings and configuring and gluing a monster. All of this work and state is not even accessible or manipulatable by an LLM - how are we supposed to be automating society by 2027 like this?"
See the post for full detail, and maybe give MenuGen a go the next time you're at a restaurant!
@karpathy@karpathy Love the “IKEA future” analogy! 🍽️ At MyBite AI we turn any menu photo into personalized dish recs, vivid food images **and full nutrition breakdowns** in seconds. Would love your take! 🔗 https://t.co/2OtUzyncVm #AI#FoodTech#NutritionTech