Arin Gopakumar

I'm a Computer Science student at UC Berkeley, specializing in artificial intelligence and machine learning. I'm currently a Software Engineer Intern at the California Department of Industrial Relations, and I spent this past summer as a Machine Learning Engineer Intern at Solugenix, where I ran systematic benchmarks of open-source speech-to-text and text-to-speech models to determine which ones were reliable enough to ship in production-grade voice agents. My research interests center on long-range graph representation learning, state-space sequence models, and applying those methods to problems in environmental forecasting and clinical prediction.

HOPPER: Learnable Hop Extraction for Linearized Graph Sequence Models

Isuru Herath, Arin Gopakumar, Sharan Sahu

arXiv preprint, August 2026

A CNN-Based Framework for Forecasting Valley Fever Risk via Dust Detection in Arizona: A Proof of Concept

Arin Gopakumar

Poster, NewInML Affinity Event, ICML 2025

WildfireSpreadBench: The Metric Decides the Model in Wildfire Spread Prediction

Arin Gopakumar, Marco Pannozzo

In preparation