Webinar June 25, 2026 · 6am PT & 11am PT 

Hot starts and batch inference: what's new in Runpod Serverless

Learn what's new in Runpod Serverless, straight from the product team. We'll cover how we shortened cold start times, batch inference, and a deployment on a 24GB Flash worker. 

30 min Demo + Q&A
Live | On demand

Speakers

Greg Wester

Greg Wester

Director of Product

Greg leads the Serverless product at Runpod. He shipped MIG, FlashBoot, and batch inference, which means he's responsible for most of what's on the agenda today. When he's not thinking about cold start latency, he coaches performance driving.




How we shortened cold start times.
Sub-second restarts via FlashBoot: what pausing workers instead of stopping them actually looks like.
FlashBoot and how hot starts work.
What's happening under the hood when a worker stays warm, and what that lets you build that wasn't possible before.

What batch inference looks like.
No standing workers. No idle burn. Built for the jobs where nobody's watching a spinner.

Live deployment.
24GB worker, Flash, cold to running. On stage.