Arjun Mehta

I started this project after struggling to visualize momentum conservation in my AP Physics C class. The textbook diagrams felt static, so I coded my own simulator to watch collisions unfold in real time. When my teacher saw it, she asked to use it in her lectures — and that feedback loop of building something useful drove me to keep improving it.

Preparing for AP Computer Science A, I realized I learned algorithms best by seeing them run. I built SortViz as a study tool for myself, then shared it with classmates. It evolved into a polished app when I decided to use it as my CS project submission. The process taught me React state management, animation timing, and deploying to production.

Fourier series clicked for me only when I saw the animation of partial sums converging. I wanted to recreate that 'aha' moment for others, so I built a tool where you can drag sliders to add harmonics and watch a jagged square wave emerge from smooth sine curves. The math behind rendering real-time waveforms at 60fps pushed me to optimize my JavaScript significantly.

My father works in finance, and I grew up hearing dinner-table conversations about how markets react to news. I wanted to test whether NLP could quantify that intuition. The project taught me about transformer architectures, data pipelines, and the humbling reality that markets are far more complex than any single model can capture — but the 67% accuracy was still better than random.