TouchGrass

AI-powered fitness and social platform with Grassy, your motivating mascot, cheering you on every step of the way!

Comment

Inspiration

During the COVID-19 lockdown, many of us felt stuck indoors, disconnected from others, and unmotivated to stay active. We missed going outside, seeing friends, and experiencing the simple joys of movement. "TouchGrass" started as a joke between us — a reminder to go outside and reconnect with the world literally. That idea stuck and evolved into something bigger: a fitness app designed to encourage real-world activity, meaningful community, and genuine moments, just like the ones we missed during the lockdown.

What it does do?

TouchGrass is a fitness and wellness app designed to help people reconnect with the real world through movement, community, and authenticity. It does three main things:

AI Fitness Coaching:

Suggests personalized activities based on your mood, goals, and schedule.

Adapts over time to better support your fitness journey.

Clubs for Social Motivation:

Lets users join or create clubs based on shared interests (like hiking, running, biking).

Encourages friendly challenges and group activities, both virtual and in-person.

TouchGrass Feed:

A "BeReal"-style feed where users post real, unedited moments outdoors or being active.

Promotes authenticity and celebration of small wins, not polished highlight reels.

How we built it

To build TouchGrass, we utilized HTML, CSS, and JavaScript for the frontend, creating a responsive, mobile-first design for users to easily navigate. The backend was powered by Node.js and Express.js to handle APIs and manage user data. For cloud storage, we integrated Firebase DB, ensuring fast and reliable data synchronization across devices. The AI-driven fitness tracker was built using TensorFlow, allowing the app to make sure you did your workout. We also used Firebase to enable real-time updates for the TouchGrass Feed and clubs, making the app interactive and community-driven. The entire app was deployed using Firebase, ensuring fast performance and scalability.

Challenges we ran into

While building TouchGrass, we encountered several challenges. One of the biggest hurdles was ensuring the AI verification system worked smoothly. The AI’s role is to verify that users actually completed their workouts, which required developing an accurate, yet simple, method for verification. This was tricky, as we had to make sure the system was not too invasive or complex, yet still reliable. Another challenge was designing a real-time, interactive experience for the TouchGrass Feed and clubs, ensuring synchronization across devices using Firebase DB. We also struggled with maintaining authenticity in the feed while preventing spam or inappropriate posts, which led to creating thoughtful systems. Finally, we had to make sure the app remained lightweight and performed well across all devices while offering all the features without making it sluggish.

Accomplishments that we're proud of

We’re incredibly proud of how TouchGrass turned out, especially considering the challenges we faced. One of our major accomplishments is the AI workout verification system, which functions seamlessly to confirm that users complete their workouts without feeling intrusive or overly complicated. We also managed to create a highly interactive and community-driven experience, with features like the TouchGrass Feed and clubs, which foster real-world connections and motivation. The integration of real-time updates through Firebase DB was another key achievement, allowing users to stay connected and engaged no matter where they are. Finally, we’re proud of how we managed to optimize the app’s performance, ensuring it runs smoothly on both iOS and Android devices, even with all the features we packed in. Seeing the positive response from users and how the app helps them stay active and motivated makes all the hard work worth it.

What we learned

Throughout the development of TouchGrass, we learned a great deal about building a community-driven app and integrating AI effectively. One key lesson was the importance of user-centered design—we had to ensure the app was intuitive, accessible, and motivating without overwhelming users. We also learned how to fine-tune AI systems, particularly when it comes to creating a non-intrusive workout verification system that still provides real value. Real-time features like the TouchGrass Feed and clubs taught us the importance of database synchronization and handling concurrent user interactions without lag. Additionally, we gained experience in using Firebase DB for cloud storage, managing user data efficiently, and keeping everything secure. Perhaps the most important lesson was learning to balance feature-richness with simplicity to keep the app lightweight while offering valuable tools for our users.

What's next for TouchGrass

Looking ahead, we're excited to take TouchGrass to the next level by making the AI fitness assistant even more personalized and intelligent. We plan to integrate AI-powered custom recommendations, where the assistant will not only verify workouts but also provide tailored fitness plans based on individual progress, preferences, and goals. This will make the app even more interactive, offering dynamic suggestions for exercises, routines, and activities that evolve as users progress. We also aim to enhance the app's social features, like clubs and the TouchGrass Feed, by introducing more community-driven challenges, events, and real-world meetups to keep users motivated. Finally, we’re exploring the possibility of adding wearable device integration to track workouts in real-time, giving users more accurate feedback and data about their activity. With these improvements, we hope to make TouchGrass an even more powerful tool for staying active, connected, and motivated.

The firebase link's might be finicky, we ran out of credits and storage :(

Built With

Share this project:

Updates