Saurabh Tangri
Principal Software Architect (AI frameworks) -
Intel CorporationSaurabh, a Senior AI Systems Architect at Intel Corporation, brings over twenty years of experience specializing in the hardware/software co-design of transformative AI acceleration technologies. At Intel, Saurabh leads as the principal architect, spearheading numerous innovative AI technologies for CPU, GPU and NPU. Under his leadership, his team has achieved breakthrough performance enhancements for the Copilot powered experiences, Windows AI platform and Microsoft Azure.
A staunch advocate for sustainable and responsible AI, Saurabh actively contributes to the Open Neural Network Exchange (ONNX) standard and various Industry initiatives at the intersection of GenAI and Trusted AI+ML Security (LFAI).
Sessions Presenting
Unleash the superpowers of the Copilot+ PC
On Demand
Interview
Online Only
Wednesday, November 20
6:17 PM - 6:21 PM Greenwich Mean Time
StudioFP122
Join us for an introduction to Intel’s groundbreaking Copilot+ PC, a revolutionary device designed to transform how we experience AI-driven computing. In this session, we’ll explore Copilot+ PC’s core pillars that sets it apart, bringing unparalleled efficiency, robust security, developer readiness, and hybrid AI integration.
SJ
Seth Juarez
Maximize AI with all processing units
On Demand
Interview
Online Only
Wednesday, November 20
8:09 PM - 8:14 PM Greenwich Mean Time
StudioFP127
Maximize AI with all processing units (Intel AIPC is CPU+GPU+NPU and not just NPU)
Experience the future of PC supporting multiple processing units with an integrated CPU, GPU and NPU. This demo shows how next-gen PCs need a fundamental capability to run LLMs on PC. See how LLMs can be partitioned for execution across processing units and run optimally. This capability is necessary to achieve best performance without the battery drain.
Experience the future of PC supporting multiple processing units with an integrated CPU, GPU and NPU. This demo shows how next-gen PCs need a fundamental capability to run LLMs on PC. See how LLMs can be partitioned for execution across processing units and run optimally. This capability is necessary to achieve best performance without the battery drain.
SJ
Seth Juarez
Accelerating real-world AI for today and tomorrow
On Demand
Breakout
In Chicago + Online
Wednesday, November 20
9:45 PM - 10:30 PM Greenwich Mean Time
BRKFP377
AI is revolutionizing how we work and play from everyday shopping experiences to scaling AI innovations across the enterprise. Learn how advancing hardware, software and solutions can help accelerate your AI initiatives from AI PCs to the data center and cloud. This expert panel will explore real-world use cases and demonstrate how AI can be securely and efficiently deployed with the power, flexibility and performance you need today and how advances in AI will help drive solutions in the future.
