Skip to content
MeetupIn personPast event

AI/ML Conversations Meetup: Smart CPU Offloading for Scalable LLM Inference

Tools & platforms

This event has ended

This listing is kept as an archive. Explore upcoming events to find your next one.

See upcoming events in New York
About this event

What to expect

In-person meetup at Capital One Flat Iron office. Bhairav Phukan presents “NEO: Unlocking Scalable LLM Inference with Smart CPU Offloading” — a technical talk on CPU offloading techniques to scale LLM inference, followed by networking and pizza. Bring photo ID for building entry.

Running an AI event in New York?

List it here, free.

Meetups, workshops and conferences go in front of thousands of builders after a quick review.

List your event