Serving Models at Scale with Red Hat AI Enterprise (AI510)
- Código del Curso AI510
- Duración 2 días
Otros Métodos de Impartición
Salta a:
Método de Impartición
Este curso está disponible en los siguientes formatos:
-
Cerrado
Cerrado
-
Clase de calendario
Aprendizaje tradicional en el aula
Solicitar este curso en un formato de entrega diferente.
Temario
Parte superiorDeploying AI models in enterprise environments demands more than a working prototype.
Organizations struggle to move AI models from experimentation to production at scale, resulting in delayed time-to-value and underutilized infrastructure investments.
This course equips teams to deploy, optimize, and operate AI models efficiently on Red Hat AI, reducing the gap between prototype and production and enabling reliable, cost-effective model serving across the enterprise.
Organizations must evaluate serving runtimes, optimize models for production hardware, manage inference at scale, and monitor deployed models to maintain performance and reliability.
This 2 days hands-on course teaches students to deploy, serve, and manage AI models at enterprise scale by using Red Hat OpenShift AI.
Students gain hands-on experience with the AI Factory framework, model serving with vLLM and KServe, NVIDIA NIM microservices, distributed inference with llm-d, workload scheduling with Kueue, and production model monitoring.
This course is based on Red Hat OpenShift AI 3.4.
Updated 01/10/2026
Calendario
Parte superiorDirigido a
Parte superiorThis course is designated for AI Platform Engineers, ML operations teams, and infrastructure administrators responsible for deploying, serving, and managing AI models at scale on Red Hat OpenShift AI.
Objetivos del Curso
Parte superiorAfter completing this course, participants should be able to:
- Evaluate serving runtimes,
- Optimize and deploy generative and predictive AI models at scale using vLLM, KServe, and NVIDIA NIM on Red Hat OpenShift AI,
- Configure distributed inference and workload scheduling,
- Monitor deployed models in production.
Contenido
Parte superiorImplementation: Implementing enterprise design patterns and structures for scalable AI operations.
Model Serving Runtimes: Evaluating and utilizing serving runtimes using vLLM and KServe.
NVIDIA NIM Microservices: Integrating optimized NVIDIA inference microservices for high-performance generative AI workloads.
Distributed Inference: Managing scalable, multi-node inference execution using llm-d.
Workload Scheduling: Optimizing GPU and compute resource scheduling with Kueue.
Production Monitoring: Tracking performance, reliability, and metrics for deployed models in production.
Pre-requisitos
Parte superior- Familiarity with Red Hat OpenShift concepts, or completion of the Red Hat OpenShift Developer II: Building and Deploying Cloud-native Applications (DO288) course.
- Basic understanding of large language models, AI/ML concepts, and model serving.
- The Developing and Deploying AI/ML Applications on Red Hat OpenShift AI (AI267) course is recommended for students who are new to the Red Hat OpenShift AI platform.
Confirmation of the correct skill set knowledge can be obtained by passing the online skills assessment at Red Hat Skills Assessment
Certificación de Prueba
Parte superiorNo certification associated with this course
Siguientes Cursos Recomendados
Parte superiorNo specific recommendation
Más información
Parte superiorOfficial course book provided to participants
