Skip to content
InfratoApp

NVIDIA · Course

Deploying and Optimizing AI Inference at Scale (instructor-led)

Deploying and Optimizing AI Inference at Scale is an instructor-led NVIDIA workshop on large-scale LLM inference from first principles, from vLLM on Kubernetes to NVIDIA Dynamo, with full-stack observability.

Open at NVIDIA ↗Common questions

Who it suits

  • Engineers who run LLM inference and want to control latency, throughput and cost.
  • Platform teams working with Kubernetes.

What it covers

  • Reasoning about inference: prefill versus decode, KV-cache scaling and memory-compute trade-offs.
  • Deploying vLLM-based inference on Kubernetes, starting from monolithic serving.
  • Using NVIDIA Dynamo for aggregated and disaggregated inference.
  • Monitoring with Prometheus, Grafana, Loki and Tempo.

Common questions about Deploying and Optimizing AI Inference at Scale (instructor-led)

What is the outcome?
NVIDIA says a principled understanding of where inference time and money go, and how architecture changes the curves.
How is it taught?
As an instructor-led workshop; NVIDIA says to work with the instructor on timing for your session.
What does it cost?
NVIDIA lists instructor-led workshops as contact-us for pricing and dates, so check its page; they change.

More like this

Keep learning on Infrato

Where this comes from

Written from NVIDIA’s own pages, read on 2026-10-14. Content, prices and availability change, so the provider’s page is the source. We are not affiliated with NVIDIA and this is not an endorsement.