# explainpls

- **Event:** [Zero to Agent: Vercel x Deepmind Hackathon SF](https://cerebralvalley.ai/e/zero-to-agent-sf)
- **When:** Sat, Mar 21 at 9:00 AM – 10:00 PM (PDT)
- **Where:** Shack15, San Francisco, CA
- **Team:** [Lars Talian Stangebye-Hansen](https://cerebralvalley.ai/u/larstalian), [Noah Lund Syrdal](https://cerebralvalley.ai/u/noahls)
- **GitHub:** https://github.com/larstalian/explainpls
- **Demo video:** https://youtu.be/Hn9EZqennF8
- **Gallery:** https://cerebralvalley.ai/e/zero-to-agent-sf/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/zero-to-agent-sf/hackathon/gallery/38

explainpls turns one photo of any physical object into an interactive learning world — not a text answer, not a quiz, not a flashcard deck. You snap something confusing, and the app produces an animated exploded-view diagram with tappable hotspots, mode-switching between subsystems, a live what-if experiment, and a grounded follow-up chat that stays anchored to the specific object you uploaded.
curiosity about how things work is universal, but every existing AI tool gives you a paragraph. A paragraph doesn't let you poke at the mechanism. We built the thing that should have existed — a mini lab you can open in 10 seconds from a phone.
The core architecture is: Gemini analyzes the photo and returns a structured LearningWorld contract (hotspots, mechanism steps, exploded scene layers, signal-flow modes, overlay paths). The frontend owns all animation, sliders, and visual state — the model never authors UI. That separation is what makes the result feel like a product rather than a hallucinated response.

---

Markdown version of https://cerebralvalley.ai/e/zero-to-agent-sf/hackathon/gallery/38. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt