Gym Buddy

• 

  Homepage

Inspiration

We wanted to make going to the gym feel less intimidating and more consistent by combining AI coaching with community motivation. The goal was to help people train with better form while feeling part of something bigger through points, leaderboards, and friendly competition.

What it does

Gym Buddy is an AI-powered workout partner that gives real-time form feedback, helps users build better workout routines, tracks progress, and rewards consistency. Users can earn points from workouts, compare progress with friends on a leaderboard, and connect a wallet to claim rewards on Solana.

How we built it

We built Gym Buddy with React, TypeScript, Vite, and Tailwind CSS. We used Gemini for form analysis and workout suggestions, ElevenLabs for voice coaching, Auth0 for authentication, and Solana + Phantom for wallet connection and reward claims. User data and workout history are stored in browser localStorage for fast prototyping.

Challenges we ran into

The biggest challenge was making the AI buddy feel truly real-time. We focused heavily on reducing AI response time and speeding up voice output so feedback arrived quickly and felt useful during workouts. Getting the full pipeline to respond fast and consistently, while still handling API delays and fallback behaviour, was the hardest part of the build.

Accomplishments that we're proud of

We shipped a full end-to-end experience: live coaching, AI analysis, workout planning, gamified competition, and crypto rewards in one app. We’re especially proud of turning complex technologies into a simple user flow that feels motivating and practical for everyday gym users.

What we learned

We learned how to combine multimodal AI, real-time frontend interaction, and web3 rewards into a single product. We also learned how important reliability, simplicity, and a smooth user experience are when integrating multiple APIs under hackathon time pressure.

What's next for Gym Buddy

Next, we plan to improve AI analysis accuracy, add social features like teams, challenges, and comments, and move from localStorage to a scalable backend. We also want to deliver more personalised coaching over time and potentially launch a mobile app.

Built With

• auth0
• elevenlabs-api
• gemini-api
• react
• tailwind-css
• typescript
• vite