MLOps and LLMOps

Entry requirements

O01 and M01; G01 for the LLM branch

What 48 hours means

Lessons

36 h

Capstone

8 h

Assessment

2 h

Remediation

2 h

Syllabus

Four modules, twelve lessons. Every lesson states what it covers and the evidence you produce for it.

• Module

  01

  3 lessons · 9 h

  

  Module 1

  Lineage

  Lifecycle and ownership · Experiment tracking · Data and feature versions

  O02Show lessonsHide lessons▾

  1. O02.01 · 180 min

     Lifecycle and ownership

     Map data, code, model, prompts and evaluation responsibilities.

     Evidence: A lifecycle register names an owner for each artefact.

  2. O02.02 · 180 min

     Experiment tracking

     Record configurations, metrics and artefact hashes.

     Evidence: A chosen experiment can be reproduced from its record.

  3. O02.03 · 180 min

     Data and feature versions

• Module

  02

  3 lessons · 9 h

  

  Module 2

  Validation

  Model and prompt registry · Offline evaluation gates · Packaging compatibility

  O02Show lessonsHide lessons▾

  1. O02.04 · 180 min

     Model and prompt registry

     Register candidate versions with approval and dependency metadata.

• Module

  03

  3 lessons · 9 h

  

  Module 3

  Release

  Staged release · Drift and quality monitoring · Feedback and labels

  O02Show lessonsHide lessons▾

  1. O02.07 · 180 min

     Staged release

     Compare shadow, canary and full rollout approaches.

     Evidence:

• Module

  04

  3 lessons · 9 h

  

  Module 4

  Monitoring

  Retraining and reindexing · Incident response · Governance evidence

  O02Show lessonsHide lessons▾

  1. O02.10 · 180 min

     Retraining and reindexing

     Trigger a controlled update with regression checks.

     Evidence:

Capstone

Build a versioned delivery pipeline for either an ML model or a RAG service. Include lineage, offline gates, staged rollout, monitoring and rollback.

Lab environment

Developer sandbox plus AI API access

Lab infrastructure is not yet provisioned, so this describes the environment the course is designed for rather than one you can launch today.