Teaching coding agents to think in GPU
Tutorial · Day 2 · Tue, Nov 10
Writing, testing, and shipping agentic skills
Session
About this session
Ask a coding agent to speed up a slow pandas job and you often get pandas back: correct, confident, and still on the CPU. The agent doesn't reach for the GPU libraries, its API knowledge is stale, and nothing checks that the GPU path was actually taken. In this hands-on tutorial, you'll fix that with agentic skills: plain markdown files that route an agent to the right library, ground it in the current docs, and validate its work at runtime. We'll take apart a real production skill for NVIDIA CUDA-X Data Science, then write, test and benchmark one together. You'll leave with a working skill and a repeatable method for any library.
Speakers
Meet the speaker
Agenda
Also on Day 2 · Tue, Nov 10
All times PT.
- 9:15 – 10:00 AMHow do agents work, and how should they work?Panel
- 10:15 – 11:00 AMEvery sandbox ends up with exceptionsPanel
- 11:15 AM – 12:00 PMBuild secure always-on local agents on AMD Ryzen AI PC using LemonadeTutorial
- 12:15 – 1:00 PMThe attack is already in your indexPanel
- 1:15 – 2:00 PMMaking a data agent trustworthy with a semantic layer and auditable memoryTutorial
Registration
Get notified for the next conference.
Registrations are live for the March 30–31, 2027 edition of the Agentic AI Conference.
Register with email
You're on the list. Watch your inbox for your Zoom link. See you November 9.
