Operationalize generative AI applications (GenAIOps)
Microsoft Learn
Last updated March 16, 2026
A path on running generative AI applications in production across the full GenAIOps lifecycle, using Microsoft Foundry, GitHub, and Python. It takes a code-first approach: planning a GenAIOps solution, managing prompts as version-controlled assets, and evaluating and optimizing agents through structured experiments with clear metrics for quality, cost, and performance. It then automates those evaluations with GitHub Actions and covers monitoring an application's performance and costs in production, plus distributed tracing with OpenTelemetry to debug complex AI workflows.
What you'll learn
- Planning a code-first GenAIOps solution
- Managing prompts as versioned assets with GitHub
- Evaluating and optimizing agents through structured experiments (quality, cost, performance)
- Automating evaluations with Microsoft Foundry and GitHub Actions
- Monitoring performance and cost, and debugging workflows with distributed tracing (OpenTelemetry)
Frequently asked questions about Operationalize generative AI applications (GenAIOps)
Who is Operationalize generative AI applications (GenAIOps) for?
Developers and DevOps engineers who want to deploy, evaluate, and monitor generative AI apps in production.
Is Operationalize generative AI applications (GenAIOps) free?
Yes — Operationalize generative AI applications (GenAIOps) is completely free to take.
What are the prerequisites for Operationalize generative AI applications (GenAIOps)?
Familiarity with fundamental generative AI concepts and services in Azure; hands-on with Python.
Do you need to code for Operationalize generative AI applications (GenAIOps)?
Yes — Operationalize generative AI applications (GenAIOps) involves hands-on coding.
Why we suggest this course
For a developer or DevOps engineer responsible for getting a generative AI app past the prototype stage, this path treats prompts, evaluations, monitoring, and tracing as engineering disciplines — versioning prompts, automating evaluation in GitHub Actions, and watching cost and latency in production. One thing to know: it builds on existing generative AI and Azure knowledge and is hands-on with Python, so it assumes you can already build the apps you are now learning to operationalize.