Harshit Chaudhary

Harshit Chaudhary

Software Engineer

About

Harshit Chaudhary is an Engineer at Jarvis Labs, who focus on GPU infrastructure and AI cloud. He works on building the infrastructure that gives developers easy access to GPU compute, so they can train and run AI workloads without managing the hardware themselves. He is working on Inference, the layer that lets anyone take a model and turn it into something people can actually use, making it easy to host, run, and chat with a model at any scale.

Type a prompt, get an answer, it feels instant. But underneath, the request quietly does a lot of work: routing, queuing, cold start, token generation, and much more under the hood, all before anything comes back. A start-to-finish look at a request and the optimizations that make it fast. 

Read More →