Fit Forge

Picture of our data enter screen
A generated workout plan tailored for a user.
Our beautiful login page
The custom home screen with a dashboard of user information.
A graph of our user statistics in our social page.

Inspiration We wanted to build a workout application that would work well for all four of our differing workout statuses. The solution we came up with was to make an app that can track your current fitness level, understand your desired goals and workout aspiration and to finally help you reach those goals. We wanted to build this to be more able to help all of us live healthier and happier lives and to be more able to help others do the same.

What it does Fit Forge will first asks the user to make an account and to set all the necessary information in the User Info tab. Once the user has entered the important information, they can either go into the workout focused "lift lab", or the diet maker section "fitness fuel". Within the lift lab, the user can add or remove from a list of specific workouts that they want to be incorporated into their routine. Once they are satisfied they can then generate a custom made workout plan for themselves that will be automatically saved into the lift lab for future reference. The fitness fuel page works similarly to the lift lab, but it is meant for making the perfect diet. Our application takes from the users statistics to find their basic metabolic rate and uses ai tools to analyze this rate, and their current goals, to make a tailored diet to help them achieve their goals.

How we built it Using Python and Streamlit, we integrated MongoDB for data storage and we used a Gemini open source api to connect with a powerful chatbot to train for each user. Building a multiple page application with streamlit was difficult, but we found some clever solutions to mitigate the effect of the technologies. The good side of streamlit is that the moment we finished the web application, the mobile component was already 100% finished as well.

Challenges we ran into We ran into a couple of major roadblocks working with mongodb and the api we used. It was very difficult to host the website online because of complications with the secrets information and library imports. Another issue we ran into while testing is that we reached the limit for our api calls in one day so our application stopped working and we couldn't figure out why.

Accomplishments that we're proud of We successfully built a functional prototype with real-time recommendations and an easy-to-use interface, leveraging the power of AI for fitness. This is an application that all of us feel comfortable using and that we would see real results if we do so. Not only did we make a product, the product is valuable to us and will be for others too.

What we learned We gained deeper knowledge of AI integrations, MongoDB management, and the challenges of building a scalable fitness app. We learned a lot of the finer details about using github to share a workflow with others and a lot about hosting services

What's next for Fit Forge We’re planning to enhance the AI-powered recommendations for even smarter, more personalized user experiences, and to make a live chatbot that will be available to answer user specific questions about the site or other workout related topics.