I just cracked the code on AI animation ads 🤯
every one of these cost me 12 cents and under 20 seconds to make
Gemini Omni + Claude Code is f*cking wild.
i built a system that turns any product into an end-to-end AI animated ad in ANY style you want.. claymation, pixar-style 3D, anime, paper cutout, lego, you name it
you drop in your product photo and a one line pitch and it handles everything end to end. images, video, voiceover, music, captions, all generated with AI inside Claude..
AI animation ads are crushing on Meta right now and everyone still assumes you need an expensive editor with a 3-week turnaround time to make one. you don't. you just need the right workflow.
Im making full end to end high quality AI animation ads in under 3 minutes with this system
and they work just as hard organically on TikTok, reels and shorts
so i packaged the whole thing up…
here's what you're getting:
> every prompt i used across all 8 steps
> the exact tool stack + the order to run it in
> style recipes for claymation, pixar, anime, wes anderson, retro cartoon + CGI
> one system that works for paid Meta AND organic
built across Claude Code and Gemini Omni
RT + reply 'STYLES' and i'll send it over (must be following so i can dm)
Teaching AI
- An AI literacy app for kids ages 6-16, teaching them about artificial intelligence related concepts ranging from neural networks, and machine learning, to prompt engineering and multi agent orchestration. All through unique labs, puzzles, and gamified content. Currently sitting at a fully developed front end, back end, 11 labs and 42 fully functioning, interactive games all coded with age band differentiation for various learning and education levels.
SparkForge - Labs (SPARKING curiosity, and FORGING skills with AI)
Hi Devansh,
Working through all UI, UX, and full game development currently, but I would love feedback on my progress thus far. Any insight is welcome, please be critical, but polite!
I am developing a web application to teach children about Artifical Intelligence and related Ai Lexus through gamified learning content! It consists of 42 games thus far within 11 lab environments covering content from basic prompt engineering, to neural networks, and multi agent orchestration.
My project: https://t.co/dMADx1bFrs
Working through all UI, UX, and full game development currently, but I would love feedback on my progress thus far. Any insight is welcome, please be critical, but polite!
Web app with 42 games/activities teaching children about AI with content ranging from machine learning to Agentic harness engineering. It’s rough but it’s finally coming to fruition!
My project:
https://t.co/dMADx1bFrs
SparkForge-Labs : Web Application Teaching children about AI through a gamified learning platform with 10 labs and 42 games. Covers a wide range of topics. To name a few: machine learning, neural networks, harness engineering, context engineering, prompt engineering all the way to creating LLMs and multi agent systems.
It has an integrated AI content generation system that continuously monitors key sources for advancements in the AI industry, then creates new content ideas (UX, Content, UI, databases, etc) funnels those through an admin dashboard for approval then fully creates, tests, and deploys the new content.
Currently 90% built. Running final audits currently. Expecting to ship full app in next 1-2 months.
Phenomenal Article and definitely worth the read.
Whatever you choose to do with AI right now whether it be engineering, building, investing, just choose something, learn fast, apply fast, ship fast, - iterate, modify and repeat.
Now is the time.
This is quite literally my favorite article I've ever posted on this account.
And arguably, the most important one you'll ever read, too.
The complete guide to investing in and capitalizing on the AI gold rush.
Don't just bookmark it - actually read it:
DAY 34: SparkForge Game Content Auditing.
*NEED FEEDBACK FROM KNOWLEDGABLE PEOPLE* Any feedback is greatly appreciated!
One of SparkForge’s 5 flagship games is Neural Builder — and it’s the most technically ambitious game in the entire platform.
Neural Builder lives in Lab 3: “The Brain Inside.” Kids will literally construct a neural network node by node.
-> On desktop, it renders as a full 3D rotatable network using React Three Fiber — glowing spheres connected by light beams, OrbitControls for rotation, around 20,000 triangles in the scene.
But here’s the design decision that matters: every game in SparkForge serves three age bands simultaneously.
Band A [PLANNED NOT YET DEVELOPED] (ages 7–9, “Explorer”) gets guided play with simple language. The neural network is visual-first : drag colorful nodes, watch them light up, see the output change. No jargon.
Band B (ages 11–13, “Adventurer”) gets scenario-based challenges. “Build a network that can tell cats from dogs.” Technical terms appear — “weights,” “layers” — but with inline definitions.
Band C (ages 14–16, “Innovator”) gets real technical language and deeper concepts. Actual discussion of backpropagation. Code-adjacent explanations. The same game, but the entire content layer adapts.
This pattern applies to nearly all of the games. Three versions of every piece of content. That’s part of what will make this project so unique — it’s not 35 games. It’s closer to 90+ learning experiences wearing 35 game interfaces.
Neural Builder is a Flagship tier game — full 3D on desktop, CSS fallback on mobile. The useIsMobile hook gates the R3F canvas entirely below 768px.
HELP ME OUT!
How would you explain neural networks to a 7-year-old?
What are your thoughts on unique age appropriate game content?
* Screenshot from mobile version for brief visual detailing. *
Day 33: Developing SparkForge - Labs
I just finished debugging the last Supabase RLS policy in Stage 3 — auth, layout shell, and the persistent chrome station frame are all locked in.
Here’s what’s wild about where I am right now:
78 documentation files written. 48 locked design decisions across 3 formal checkpoints. 14 enhancement documents totaling 34 part files. 35 games designed, content-mapped, and tier-assigned.
Currently Stages 1 through 3 are fully developed, debugged, and tested.
The foundation layer (Next.js 14 App Router + TypeScript strict mode), the database and API layer (Supabase PostgreSQL + Row Level Security), and the auth + layout shell are done.
Tomorrow I start Stage 4 — Core Pages and UI. The dashboard, the lab selection interface, the profile system. This is where the Laboratory Control Station metaphor starts becoming real on screen — chrome bezels, LED status rims, the aurora shader background.
33 days of building a platform that teaches kids AI through 35 interactive games across 10 themed labs.
As always shoutout @AnthropicAI and @claudeai for literally assisting me in brining my dreams to fruition. It’s unreal what we can be capable of now.
Stage 4 is where the station powers on. Stay tuned!
@ihteshamali As I know you have developed some seriously awesome web apps, have you had any success with the webapp-testing skill?
Interested to hear your or anyone’s insights on this one in particular.
https://t.co/8Kpcb9r7sU