Summary
Harsh Gondaliya is a production engineer specializing in AI infrastructure and edge GPU systems, with 8 years of experience and 3+ years focused on deploying and optimizing GPU-backed inference in production. At Meta he repeatedly earned top performance ratings for projects that cut Meta AI Voice end-to-end latency by hundreds of milliseconds through smarter routing, edge deployments, and protocol simplifications. He designed global autoscaling and capacity primitives for 120,000+ GPUs across 50 metros that unlocked $8M+ annual savings and near-universal adoption without service disruption. Comfortable across low-level networking, data-plane programming (P4/NetFPGA), and higher-level orchestration, he combines research rigor from UC San Diego with hands-on production impact. Outside core infra work he has built active Internet measurement pipelines and mentored contributors to expand global vantage points, showing a knack for scalable tooling and reproducible experiments.
8 years of coding experience
2 years of employment as a software developer
University of California, San Diego
Higher Secondary Education (CBSE), Higher Secondary Education (CBSE) at Atmiya Vidya Mandir - International Residential School for Boys
Bachelor of Technology (B.Tech.), Computer Science & Engineering, Bachelor of Technology (B.Tech.), Computer Science & Engineering at SRM IST Chennai
English, Gujarati, Hindi, Chinese