-
Brief Screenshot of the UI
TradeLab was inspired by the idea that learning about investing should be accessible to everyone. Financial markets can feel intimidating to beginners, especially when learning involves real money and unfamiliar terminology. I wanted to create a platform where people could explore trading concepts, practice with virtual funds, and build financial knowledge in a low-risk environment.
TradeLab combines simulated trading, beginner-friendly educational courses, a trading journal, and an AI-powered tutor in one platform. Users can practice trading strategies, reflect on their decisions, and learn about financial concepts with guidance from an AI assistant. The goal is to make learning about investing more interactive and approachable.
TradeLab uses a web-based frontend connected to Python Flask backends. We use a locally hosted Qwen language model through llama.cpp to power the AI tutor, reducing reliance on external AI APIs. A financial knowledge base supports the tutor with relevant educational information.
The project brings together frontend development, backend APIs, authentication, data persistence, local AI inference, and information retrieval. We also worked on integrating separate parts of the application into a cohesive experience.
One of our biggest challenges was integrating independently developed components. Combining the frontend, account system, and AI tutor required careful attention to Git branches, API communication, dependencies, and preserving existing functionality.
Running an AI model locally presented additional challenges, including model setup, hardware limitations, backend connectivity, and generating reliable responses. We also needed to consider how to provide useful educational guidance without presenting AI-generated information as guaranteed financial advice.
Building TradeLab helped us better understand how full-stack applications connect frontend interfaces, backend services, databases, and AI models. We gained practical experience with Git collaboration, API integration, local model deployment, and retrieval-augmented generation.
Most importantly, we learned that building an AI-powered product involves more than getting a model to respond. It requires thoughtful user experience, reliable infrastructure, appropriate safeguards, and a clear understanding of the problem the technology is meant to solve.
We want to continue improving TradeLab by refining the AI tutor, expanding educational resources, strengthening account and data management, and making the simulated trading experience more useful for beginners. Our long-term goal is to make financial education more approachable and help users develop informed, responsible investing habits.
