# Loop

- **Event:** [Meta Global AI Developer Hackathon](https://cerebralvalley.ai/e/meta-ai-developer-hackathon)
- **When:** Oct 26, 12:00 PM – Nov 4, 9:00 PM (EST)
- **Where:** Online
- **Team:** [Divya Kamila](https://cerebralvalley.ai/u/divyak07)
- **GitHub:** https://github.com/divya0702/loop
- **Demo video:** https://youtu.be/Lzml8PeXlhA
- **Gallery:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3724

Loop 
Your prep, on loop.

Loop turns your own notes and code into a role-specific preparation plan for interview, then helps you practice, receive feedback, and improve, all in one workspace.

The problem I wanted to solve

While preparing for interviews, I had no shortage of material. I had personal notes in OneNote, bookmarked tutorials, coding problems, and resources scattered across websites.

What I lacked was direction. Which material mattered for the role I was targeting? What should I practice today? And after practicing, what should I improve?

I built Loop to connect those pieces into a repeatable preparation cycle.

How Loop works

Start by bringing your notes, documents, and code into Loop and selecting your target role. From there, you can:

Build a preparation plan: Generate relevant topics, drills, and a five-day schedule from your material.

Explore your knowledge: Turn notes into an interactive mindmap, explore connections, and launch a drill from a concept.

Practice with feedback: Answer technical, behavioral, or leadership questions and receive scores, actionable feedback, a model answer, and follow-up questions.

Rehearse an interview: Run an adaptive mock interview that responds to your answers and ends with a structured assessment.

Reason through code: Explore AI-generated execution simulations with step-by-step variable states and potential gotchas.

Fill knowledge gaps: Request suggested resources, save useful ones, and track your reading.

How Meta powers Loop

Meta Muse Spark is Loop’s intelligence layer. Every AI feature makes a live call through Meta Model API: planning, question generation, answer feedback, adaptive interviewing, interview assessment, mindmap generation, resource suggestions, and code simulation.

These calls run behind two Supabase Edge Functions, with the API key held server-side. React and TypeScript provide the interface, while Supabase handles authentication, saved data, and access controls. I also used Muse Code during development.

What makes it personal

Loop builds on my existing interview-preparation foundation. This entry focuses on the connected, Meta-powered workflow: turning my own material into a plan, making that plan actionable through practice, and using feedback to guide the next attempt.

The goal is to make every preparation session lead to a clear next step.

Try Loop: https://divya0702.github.io/loop/#

Your material in. Interview readiness out.

## More from Meta Global AI Developer Hackathon

- [JoinWitness](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3721)
- [Prestige](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3722)
- [Vector Minds](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3723)
- [Portravo](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3725)
- [Alex Red](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3726)
- [Raala](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3727)

---

Markdown version of https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3724. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt