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

We are delighted to announce that our upcoming meetup is scheduled for Wednesday, November 12th. We invite you to join us at the Capital One Flat Iron office to discuss the latest advancements in Machine Learning and Artificial Intelligence. There will also be an opportunity to network and enjoy some pizza.

​

Details

Agenda:
5:30PM Doors open
5:30PM - 6:15PM Reception, networking
6:15PM - 7:15PM Bhairav Phukan, NEO: Unlocking Scalable LLM Inference with Smart CPU Offloading
7:15PM - 8:30PM Further Networking & get-together in the nearby bar.

!!! IMPORTANT !!!
- Please provide your full legal name at the time of RSVP, as it is necessary to enter the building and please ensure to bring a valid photo ID.

This event is hosted at the Capital One Office! Enter through 11 W 19th Street main entrance. Show your photo ID at the front desk and proceed to the 3rd floor.

Speaker

Bhairav Phukan is a Master’s student in Computer Engineering at Columbia University. With a strong foundation in AI and embedded systems, Bhairav focuses on building scalable, socially impactful technologies at the intersection of deep learning, human-computer interaction, and assistive tech.

He is a winner of multiple international hackathons, including the prestigious 2022 ASME-CIE Hackathon, where he led his team to 1st place against MIT Boston, becoming the sole undergrad student to defeat PHD students in the conference organised by American Society of Mechanical Engineers.

Bhairav has also contributed to impactful academic research. His first publication, in collaboration with IIT Madras and the Director General of Police, Tamil Nadu, examined the impact of COVID-19 on maternal healthcare using a novel deep learning architecture combining Feedforward Neural Networks and Auto-Regressive models. This work was published in PLOS ONE and used Neural Prophet for causal time-series analysis.

In his second research project, Bhairav developed a YOLOv8-based pothole detection and depth estimation system that has been integrated into a patent-pending smart assistive device for the visually impaired. The device also includes additional features such as LLM-enabled SPO2 monitoring for broader accessibility and healthcare use.