//

Day 1 · Mon Nov 30

·

Midday

·

25 min

AI is becoming its own performance engineer

AI is becoming its own performance engineer

AI is becoming its own performance engineer

Talk

// 01 / About this session

AI is changing how engineers optimize AI systems, from AI-assisted GPU kernel development to more efficient model architectures and inference.

Simran shows how AI-driven optimization techniques improve performance, latency and efficiency, with practical lessons on working within hardware constraints and optimizing real-world AI workloads.

What you’ll learn

How AI is changing the way engineers optimize AI systems

Where AI-driven optimization improves performance, latency and efficiency

Practical lessons on hardware constraints and real-world AI workloads

Skills you’ll practice

Operating in production

Designing system architectures

Scaling and operating in production

// 02 / Speakers

Simran Arora

Simran Arora

SVP of Research, Together AI

SVP of Research at Together AI, leading its Frontier Performance team. Stanford PhD, incoming Assistant Professor at Caltech, and contributor to ThunderKittens and KernelBench.

AI DEV · BY DEEPLEARNING.AI

Join us at AI Dev

Meet the engineers shipping AI to production, and leave ready to do it yourself.