# Gapseeker Labs

- **Event:** [Gemini 3 Hackathon London](https://cerebralvalley.ai/e/gemini-3-hackathon-london)
- **When:** Sat, Dec 13 at 9:00 AM – 10:00 PM (UTC)
- **Where:** CodeNode, London, UK
- **Team:** [Dave Cheng](https://cerebralvalley.ai/u/Masterasnackin), [Claire Sheng](https://cerebralvalley.ai/u/Claire356), [Guiru Lin](https://cerebralvalley.ai/u/Guiru)
- **GitHub:** https://github.com/MasteraSnackin/Gapseeker
- **Demo video:** https://youtu.be/ipYlWF2LnFM
- **Gallery:** https://cerebralvalley.ai/e/gemini-3-hackathon-london/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/gemini-3-hackathon-london/hackathon/gallery/27

Gapseeker turns educational videos into actual learning systems instead of just something you watch and forget. It uses Gemini 2.5 Flash to break down a video, figure out what's being taught and when, then builds quizzes, lesson plans, and study materials around it.

From a single video, Gapseeker can:

Map where key concepts appear along the timeline

Generate quizzes tagged by cognitive level (recall to synthesis)

Create lesson plans for different durations (5-min review, 15-min deep dive, 45-min lecture)

Produce flashcards with both technical explanations and simple analogies

There's also a voice-enabled tutor and a "Think Tank" mode where two AI personas—The Academic and The Simplifier—discuss concepts from different angles and converge on a clear explanation.

The app works with local video files and YouTube links, built with React 19, TypeScript, Tailwind CSS, and Gemini 2.5 Flash multimodal API.

---

Markdown version of https://cerebralvalley.ai/e/gemini-3-hackathon-london/hackathon/gallery/27. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt