← Journal · · Announcement
Triton inference platform: talk at TechDay
Hey everyone! This Thursday I'll be giving a talk about our new product, an inference platform built on NVIDIA Triton Server, at the TechDay conference.
You can watch online if you register via the link
I'll talk about the model lifecycle, why building your own inference from scratch isn't a great idea, and how our platform is built
I'll cover autoscaling, how we get rid of downtime with canary deploys, and how to build an inference graph with Ray!
Also, if you're coming offline, I'll show a live demo of the platform, how to use it)
We'll also have a booth where we'll talk about what's up with AMD and how it competes with NVIDIA in the ML space. Register and join us!