“I See Dead Pipelines...Without CI/CD:" The Gen AI Easy Button for Fine-Tuning, Evaluation, and Inference With NVIDIA NIMs and NeMo Microservices

"Dead pipelines" — abandoned scripts that once ran parts of a workflow — plague LLM development, causing errors, delays, and wasted resources. Learn how to automate the entire LLM life cycle using CI/CD pipelines, Kubernetes, NVIDIA NIMs, and NeMo Microservices in a cloud-agnostic approach. Discover how to streamline development, fine-tuning, evaluation, and deployment of LLMs, ensuring seamless integration, validation, and deployment of updates.

In this hands-on course, you'll:
Design and build a cloud-agnostic LLMOps CI/CD pipeline on Kubernetes for scalable and flexible Gen AI app deployment.
Utilize Argo CD for declarative continuous delivery and Argo Workflows for managing complex LLM workflows.
Leverage NVIDIA NIMs and NeMo Microservices in the end-to-end LLMOps pipeline.
Learn how to improve development cycles, accuracy, and reliability with automated fine-tuning and continuous evaluation.
Prerequisite(s):

Kubernetes basic knowledge.
DevOps tools basic knowledge like ArgoCD.