Current¢ was born out of a shared frustration: half our team are international students, and we’ve all felt the hassle of transferring money across borders. At a time when students should be focused on adjusting to a new environment, dealing with unpredictable exchange rates adds unnecessary stress. We wanted to create something that made this process clearer, smarter, and easier to manage.

Our app provides real-time exchange rates between any two currencies, paired with a graph of the past month’s trends to help users judge whether the current rate is favorable. It classifies rates as good, average, or bad based on recent data and includes a “notify me” option so users get alerts when rates improve.

We built everything in Python using the Shiny GUI framework, converting it to HTML and WebAssembly to make it lightweight, browser-based, and server-free. Despite facing challenges with limited documentation and package constraints in WebAssembly, we powered through with a working prototype that captured our vision. Current¢ went on to win three awards at WHACK 2023: Most Readily Employable, Best Financial Hack with Capital One, and Best Domain Name from GoDaddy Registry.

As part of catalan.ai’s broader effort to support smarter, more sustainable energy systems, this project combines tree-based regression models with Large Language Models (LLMs) to improve electricity demand forecasting and generate interpretable, actionable insights for energy management. By blending the predictive power of traditional machine learning with the reasoning capabilities of LLMs, we aim to make energy data more transparent and usable for real-world decision-making.

Using operational data from ERCOT—including real-time fuel mix, renewable energy generation, and locational marginal prices—we trained regression models such as Random Forests to forecast demand with high accuracy. LLMs were then used to interpret the model outputs, answering key questions around pricing trends, volatility, and optimal load balancing. This hybrid approach supports energy providers in making informed, forward-looking decisions that align with both operational efficiency and sustainability goals.