Inspiration

The idea for MindCoach came from my frustration with generic online courses that repeat known material. I wanted to create a tool that adapts to each learner’s level, inspired by tutoring friends and seeing their time wasted on basics.

What it does

MindCoach offers personalized learning paths where users select a subject (e.g., Python), take a survey to assess their knowledge, and receive 10 tailored markdown lessons in users/<user_id>/<subject> folders, skipping mastered topics. It features a pay-per-subject model.

How we built it

Using Kiro’s Vibe mode, I brainstormed the concept, then used Spec mode to define requirements. The app was built with React and Tailwind for a responsive UI, and Flask with SQLite for the backend. Kiro’s code generation created the survey UI, and hooks automated folder setup.

Challenges we ran into

A key challenge was ensuring the survey accurately gauged knowledge to skip topics, requiring multiple iterations. Another was optimizing the UI for all devices, which took trial and error with Tailwind’s responsive design.

Accomplishments that we're proud of

We’re proud of the seamless user-specific lesson delivery and the responsive design that works across desktop, tablet, and mobile. The pay-per-subject model’s implementation also feels like a unique, user-friendly innovation.

What we learned

I learned to structure projects with Kiro’s Spec mode, integrate React with Tailwind effectively, and use hooks for automation. Balancing survey accuracy with lesson personalization was a valuable lesson in iterative development.

What's next for MindCoach

Next, we plan to add AI-driven lesson adjustments based on user progress and expand subject offerings. A community feature for peer learning is also on the horizon.

Built With

Share this project:

Updates