AI Infrastructure, Built in the Open
Choose one half-day experience - or attend both for a full day of technical learning and open-source contribution.
Morning: AI Inference Technical Deep Dive
Technical sessions and live demonstrations covering deployment, load testing, scaling, and platform operations. The morning program does not include individual attendee labs.
Afternoon: DSX OS Contributor Workshop
A hands-on working session where participants collaborate with maintainers, work directly in open-source repositories, and begin or advance a contribution.
Schedule at a Glance
November 9th
NVIDIA AI Inference Technical Deep Dive
NVIDIA AI Inference Technical Deep Dive
Coffee and networking with NVIDIA engineers and maintainers.
Introduce the platform stack and demonstrate provisioning a Kubernetes/GPU foundation, deploying an open model, and issuing a live inference request.
See how to connect a coding agent, trace inference, build repeatable load tests, scale GPUs, and accelerate model loading.
Explore replica scaling, routing strategies for cache efficiency, and the separation of prefill and decode workers across a cluster.
Learn how shared GPU capacity, workload checkpoint and restore, and emerging research problems are addressed in multi-team environments.
Optional working lunch to get set up as an open-source contributor
DSX OS Contributor Workshop (Half-Day)
DSX OS Contributor Workshop (Half-Day)
Fast overview of DSX OS projects and contribution routes.
Hands-on work in the repositories—claim an issue, open a draft PR, run CI, get live maintainer review
Rapid recap and recognition of sprint accomplishments, then continued PR review and CI troubleshooting for unfinished work
Small-group architecture discussions. Bring your org's requirements and integration ideas
Open networking to connect with maintainers and partners, closing with a recap and each attendee's concrete next action
Venue
Wasatch Ballroom- 2nd Floor
Radisson Hotel Salt Lake City Downtown
215 West South Temple, Salt Lake City, UT, 84101
