Nilesh Agarwal
CTO
Actions
Nilesh Agarwal is the Co-Founder and CTO of Inferless, the serverless GPU inference platform that lets teams deploy and scale machine learning models with near-instant cold starts. He co-invented a patented graph-based anomaly detection system (US 11,928,629 B2) that uses optimized graph encodings for static code analysis. During his tenure at Amazon, he helped develop ML-driven ad-optimization features for Amazon Ads, contributing to the platform’s growth.
Scaling ML/AI Application Faster with Three-Tier Fuse Storage
Large language models (LLMs) often span tens or hundreds of gigabytes, and loading them repeatedly across distributed GPU clusters can turn storage into a critical bottleneck. In this talk, we’ll introduce a three-tier storage architecture—Hot (local NVMe SSDs), Warm (intra-cluster file sharing), and Cold (cloud object storage)—that balances performance, cost, and durability to feed your GPUs without pause.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top