VectorBench

Inspiration

After seeing a stock trading professional run multiple bots with 20 simultaneous trading strategies we wanted to replicate something in the open source world that could be used by aspiring bot traders and market enthusiasts. We wanted to build a tool that could allow various stock trading bots to compete head to head in simulated environments using completely real world events and stock data.

What it does

VectorBench is a Market Trading bot optimization tool that recreates previous market trends and determines a bots performance and accuracy while using generative AI to classify any anomalies that the bot has based on relevant data.

How we built it

We used Next.js, Tailwind, Python, and Flask for the website. SciKit, Pandas, Numpy, and MatPlotLib for the classification and Deepseek reasoning models for guided help.

Challenges we ran into

Gathering historical financial news and accurate reporting data for proper forecasting was arguably our hardest challenge.

Accomplishments that we're proud of

We're extremely proud that we were able to create such a intricately designed tool to assist developers and such a beautifully designed website.

What we learned

Since we had multiple beginners, most of us learned how to navigate through our first hackathon. And others overcame challenges becoming better programmers.