PortfoliGO

Drafting System
Stock selection from free market
Stock Growth Analytics
A weekly matchup

Inspiration

While brainstorming for a fresh idea, a team member received an ESPN notification, sparking the concept of a fantasy-style game. We chose stocks as the focus because they offer real-world relevance, practical financial insight, and the excitement of competition blending entertainment with meaningful market engagement.

What it does

PortfoliGO is a fantasy stock-trading platform that gamifies investing by letting users create and join leagues where they draft and manage stock portfolios instead of sports players. Current market data coupled with machine learning models power value, risk, and growth scores, and the best-performing portfolios rise to the top.

How we built it

To bring this to life, we built a modern React and TypeScript front end on the Vite framework for speed and optimized performance. Supabase provides secure backend services, authentication, and real-time database management. Our machine learning models, developed in Python to score stocks on value, risk, and growth, are deployed through a Flask API. This cleanly integrates the ML logic with the app so model outputs and collected data flow seamlessly into Supabase and stay updated in real time.

Challenges we ran into

During development, we faced many unexpected challenges. Extreme outliers in terms of market data such as GameStop or AMC caused problems to our machine learning models. These stocks have extreme outliers with datapoints such as 1400 % EPS growth. We solved this by building anomaly limiters so metrics beyond defined thresholds taper off or cap, keeping scores reliable despite market hype. Integrating the Python ML pipelines with the front end was another major hurdle. We addressed it by creating a robust Flask API layer to handle secure, structured requests and automatic database updates, creating a scalable bridge between machine learning and the user interface.

Accomplishments that we're proud of

We’re proud of how we turned complex financial analysis into an engaging fantasy experience. Our Random Forest and XGBoost models combine a multitude of financial factors to generate rich, representative stock scores. The interface captures the excitement of fantasy sports while preserving analytical depth, and the seamless integration of risk, growth, and value metrics makes the platform both fun and educational.

What we learned

This project also gave us hands-on experience with the full development lifecycle and system-wide integration skills rarely covered in traditional classes. Tackling real-time data flows, multi-service orchestration, and scaling considerations forced us to problem-solve on the fly and taught us how to connect every layer of a complex application.

What's next for PortfoliGO

Looking ahead, our next step is to develop a mobile app to reach a wider audience. Built with React Native and backed by Supabase, it will broaden accessibility and engagement. From there, we plan to scale rapidly, optimizing infrastructure, leveraging cloud hosting, and implementing automated deployment and monitoring to handle surging traffic and data growth. With this trajectory, not only could PortfoliGO skyrocket, but so could your portfolio.