CarMind+

• 

  App Guide Page 1

• 

  App Guide Page 2

• 

  First Page

• 

  After Diagnosis

• 

  Sample Diagnosis

• 

  Sample Diagnosis

• 

  Sample Diagnosis

Inspiration

As an international student in the US, I noticed that almost all my friends own cars and have experienced unexpected breakdowns or issues at least once. Many of them found car repairs stressful, expensive, and confusing—especially when they didn’t know what the problem was or whether the mechanic’s recommendation was really necessary. This inspired me to build CarMind+, a tool that helps anyone, regardless of technical background, to diagnose and solve common car problems themselves, save money, and feel more confident about car maintenance.

What it does

CarMind+ is an AI-powered assistant that guides users step-by-step through the process of diagnosing and repairing common vehicle issues. Users can describe their car’s symptoms, and the app provides possible causes, a ranked list of solutions, and detailed DIY repair instructions with estimated cost, time, and difficulty for each step. The goal is to empower people to make informed decisions about their car and even attempt simple repairs themselves.

How we built it

I built CarMind+ using React, TypeScript, and Tailwind CSS to create a clean and user-friendly interface. The diagnosis logic currently uses mock data to simulate realistic car problems and solutions, making it easy to test the app and demonstrate its value without needing real vehicle data or hardware. The development workflow was mostly powered by Bolt, and the chat-based interface allowed me to rapidly prototype, fix bugs, and improve features in real-time.

Challenges we ran into

The biggest challenge was making complex automotive information accessible and understandable for beginners. It was also difficult to design a UI that could show step-by-step guides, tool lists, cost estimates, and safety information without overwhelming the user. Since I did not have access to a real vehicle or OBD-II data, I had to creatively use mock data to ensure the app still felt useful and realistic. Making sure that the app gracefully handled missing data or edge cases was another important technical hurdle.

Accomplishments that we're proud of

I’m proud that I was able to design and build an application that can genuinely help people like my friends—international students and anyone unfamiliar with car repair—feel more confident and capable. I’m also happy with how user-friendly and clean the interface turned out, and how quickly I could develop new features and troubleshoot problems using Bolt. This project has already sparked interest from friends who want to use the app for their own cars.

What we learned

As someone without a background in automotive engineering, I started this project with very little knowledge about cars. While building CarMind+, I had the chance to study how vehicles work, learn about common problems and repairs, and practice translating technical concepts into clear, simple instructions for other beginners. This experience not only improved my technical and design skills, but also gave me the confidence to tackle unfamiliar topics through hands-on learning.

What's next for CarMind+

Although CarMind+ currently uses mock data and does not yet provide full video guides, I am committed to continuously improving the app. In the future, I plan to add curated repair video guides for each step, provide direct links to purchase necessary parts, and integrate real diagnostic data from actual vehicles. I also want to support more car models and languages, and create a community-driven knowledge base where users can share their own repair tips and experiences. My ultimate goal is to make car care accessible, transparent, and empowering for everyone.

Built With

• bolt
• bolt.new
• mock-diagnosis-api(custom-implementation)
• react
• tailwind-css
• typescript