stardance has been extended another month! the new deadline is october 31 :)

You are browsing as a guest. Sign up (or log in) to start making projects!

princepatel

@princepatel

Joined October 15th, 2026

  • 1Devlogs
  • 2Projects
  • 1Ships
  • 0Votes
Ship Pending review

I built Prep OS, an AI-powered study workspace and syllabus tracker designed specifically for Class 12th science and competitive exam preparation across Physics, Chemistry, and Mathematics. To solve the problem of ineffective revision, I built an Error Log AI System that uses the Google Gemini API to analyze mock test mistakes, categorize error types (conceptual, calculation, or misreading), and generate targeted practice tasks. I also created an AI Schedule Maker that calculates remaining syllabus requirements and generates personalized, time-blocked daily timetables using active recall and spaced repetition. The biggest challenge was structuring Gemini API prompts to parse speech/text mistake inputs into clean JSON outputs and managing automated deployments on Render via GitHub. I am proud of creating a comprehensive tool that turns test mistakes into actionable study plans. Anyone testing the live site on Render can update chapter completion bars, log recent test mistakes into the Error Log AI to receive automated revision strategies, and run the AI Schedule Maker to generate a personalized timetable.

  1. What was challenging?
    Integrating the Google Gemini API to reliably power both the Error Log analysis and dynamic schedule generation was challenging. It required crafting precise prompt structures to convert speech/text inputs into actionable study insights, as well as managing deployment on Render with seamless GitHub backend integration.

  2. What are you proud of?
    I am proud of building a unified study workspace that goes beyond basic progress bars. Combining automated mistake analysis with smart timetable generation creates a practical, high-value tool that directly optimizes real-world exam prep workflows.

  3. What should people know so they can test your project?
    Users can visit the deployed app on Render to test all core modules: track subject/chapter progress, input recent test errors into the Error Log AI for targeted review suggestions, and use the AI Schedule Maker to generate a personalized daily study routine.

  • 1 devlog
  • 0h
Try project → See source code →

Followers

Loading…