Advanced|15 hours|36 lessons

GitOps with Argo CD

A production course on GitOps with Argo CD. The reconciliation model, repository and environment structure, app-of-apps and ApplicationSets, sync mechanics and their failure modes, secrets, multi-cluster and multi-tenant topologies, progressive delivery with Argo Rollouts, and running Argo CD as critical infrastructure. For platform engineers and SREs who own deployment.

Text-based, no videos
8 modules, 36 lessons
Lifetime access

What you'll learn

What GitOps actually requires beyond "our manifests are in Git": declarative state, pull-based delivery, and continuous reconciliation, plus the problems it does not solve
Argo CD architecture component by component, the Application object model, and the reconciliation loop in enough detail to debug it
Repository and environment layout that survives growth: monorepo against polyrepo, app-of-apps bootstrapping, and ApplicationSet generators
Sync mechanics in depth: apply strategies, sync waves and hooks, diffing and permanent OutOfSync, self-heal and prune, and how to unstick a stuck sync
Secrets in a GitOps world: Sealed Secrets, SOPS with KMS, and External Secrets Operator compared on rotation, recovery, and operational cost
Multi-cluster topologies and multi-tenant access control with AppProjects, RBAC, and SSO, designed so dangerous actions are impossible rather than discouraged
Progressive delivery with Argo Rollouts: canary and blue-green, metric-driven analysis, automated rollback, and how Rollouts interacts with Argo CD self-heal
Running Argo CD as critical infrastructure: observability, disaster recovery from Git, upgrades, and performance tuning at thousands of Applications
A capstone design exercise covering many teams, many clusters, secrets, promotion, and progressive delivery end to end

Curriculum

8 modules · 36 lessons
01

The GitOps Model

What GitOps actually is beneath the marketing: declarative state, pull-based delivery, continuous reconciliation, and the problems it does not solve.

4 lessons
02

Argo CD Architecture

The components, the object model, and the reconciliation loop in detail, including where Argo CD spends its time and where it gets slow.

5 lessons
03

Repository and Environment Structure

Laying out repositories, bootstrapping with app-of-apps, generating Applications with ApplicationSets, and promoting changes between environments.

5 lessons
04

Sync Mechanics and Failure Modes

What actually happens during a sync: apply strategies, ordering with waves and hooks, diffing and drift, self-heal and prune, and why syncs get stuck.

5 lessons
05

Secrets in a GitOps World

The one thing you cannot commit: encrypted-in-Git approaches, external secret stores, and how each handles rotation and disaster recovery.

4 lessons
06

Multi-Cluster, Projects, and Access Control

Running Argo CD across many clusters and many teams: topology choices, AppProjects as tenancy boundaries, RBAC, SSO, and self-service guardrails.

4 lessons
07

Progressive Delivery with Argo Rollouts

Going beyond a rolling update: canary and blue-green strategies, metric-driven analysis, automated rollback, and how Rollouts interacts with Argo CD.

4 lessons
08

Production Operations

Running Argo CD as critical infrastructure: observability, disaster recovery, upgrades, scale tuning, and an end-to-end design capstone.

5 lessons

About the Author

Sharon Sahadevan

Sharon Sahadevan

AI Infrastructure Engineer

Building production GPU clusters on Kubernetes. H100s, large-scale model serving, and end-to-end ML infrastructure across Azure and AWS.

10+ years designing cloud-native platforms with deep expertise in Kubernetes orchestration, GitOps (Argo CD), Terraform, and MLOps pipelines for LLM deployment.

Author of KubeNatives, a weekly newsletter read by 3,000+ DevOps and ML engineers for production insights on K8s internals, GPU scheduling, and model-serving patterns.

Ready to master this topic?

Start with the free preview lesson and see for yourself.