Conference Talk
Deploying AI Applications on Kubernetes
A hands-on guide to containerizing and deploying AI applications with scalability and observability.
KubeCon + CloudNativeCon 2023
Amsterdam, Netherlands · Apr 18, 2023
About the Talk
Model servers, GPU scheduling and bursty inference traffic stretch default Kubernetes setups. This talk shows a reference deployment for AI services — autoscaling on queue depth, GPU node pools, and tracing that follows a request from gateway to model.
Cite This Talk
Saha, B. (2023). Deploying AI Applications on Kubernetes. KubeCon + CloudNativeCon 2023, Amsterdam, Netherlands. https://iambishal.com/work/research/talks/deploying-ai-applications-kubernetes