VATSAL

MEHTA

I work at the intersection of machine learning and real-world problems — building systems that extract meaning from speech, text, images, and scientific data. Most of my time goes into deep learning, NLP, and computer vision.

I care about the full pipeline — from messy data to a model that actually works in production. Whether it is training transformers on audio, scraping and summarizing research papers, or simulating biological systems, I want the output to be useful, not just accurate on a benchmark.

Python is my primary language. PyTorch and TensorFlow are my daily tools. I am always reading papers and looking for the next hard problem to solve.