Description
This LLMOps bootcamp follows an assistant called Opsmate from a local prototype into a managed model-serving workflow. It examines token processing, caching, quantization and model selection in relation to latency and memory use. Retrieval pipelines, evaluation datasets and LoRA fine-tuning provide the basis for deciding whether a model or application change is ready to progress.
The operational sections cover packaging and signing model artifacts, Kubernetes rollout and rollback, GitOps version management and queue-based scaling. Monitoring, request gateways, budgets and restricted tool access are treated as parts of the same production workflow. The exercises are designed around local models and containers, with the stated hardware and command-line requirements retained separately.
Intended audience: Platform, infrastructure, site-reliability and backend engineers moving into LLM operations. Machine-learning engineers and technical leads responsible for deployment decisions.
What you will learn
- Measure model latency, memory use and retrieval quality.
- Build evaluation datasets and assess fine-tuning changes.
- Package model artifacts and manage Kubernetes deployments with GitOps.
- Monitor queues, traces and token usage while constraining agent tools.
Course specifications
- Publisher
- Udemy
- Instructors
- Gourav J. Shah, School of Devops
- Quality
- 1080p
- Level
- Beginner to Advanced
- Duration
- 12 hours and 30 minutes
- Subtitle
- None
- Edition
- 2026-7
Course topics
View course topics
The course topics image is temporarily unavailable.
Prerequisites
- A laptop with at least 8 GB RAM and about 40 GB available storage.
- Familiarity with command-line tools, Git, containers and Docker Compose.
- Basic Kubernetes knowledge is helpful for later sections, but not required.
Pictures
This picture is temporarily unavailable.
Introduction video
How to open the files
Keep all parts of the same edition together before extracting the first archive. Open the included course media with a compatible player.
Download resources
Choose one edition and keep its parts together. Alternate providers are available when listed.
Size
1.5 GB
Leave a Reply
Signed-in member comments appear immediately.