Learn KEDA from the basics to production readiness: pre-requisites & environment setup, history & why you need KEDA, core concepts & operator architecture, Helm setup & installation, ScaledObject & ScaledJob, scale-to-zero & activation, TriggerAuthentication, basic scalers, message queue scalers SQS & RabbitMQ & Kafka, database & HTTP scalers, custom scalers & external metrics, fallback & advanced config, HTTP add-on deep dive, security & credentials, multi-tenancy & RBAC, best practice & cost, integration with Karpenter, advanced scalers & ecosystem, troubleshooting & monitoring, performance & tuning, latest stable features v2.20, production-ready deployment, and alternative ecosystems & final reflection — 23 episodes in total.
Opening episode of the Learn KEDA series: the Kubernetes skills and event-driven concepts you must master, the tools you need to prepare, and the environment verification steps before diving into the core KEDA material.

KEDA's journey from a 2019 project by Microsoft and Red Hat to a graduated CNCF project, the problems it solves, and its comparison with pure HPA and the custom-metrics-adapter.

Opening up KEDA's architecture: the operator that manages ScaledObject and ScaledJob, the metrics server that feeds metrics to HPA, admission webhooks, plus the main CRD components and the role of 70+ scalers.

Complete KEDA installation guide: install via Helm, verify running components, configuration options via --set, tolerations, a manifest alternative for GitOps, and AKS and EKS cloud add-ons.

Two core KEDA objects: ScaledObject for continuous workloads and ScaledJob for batch workloads, complete with key fields such as pollingInterval, cooldownPeriod, minReplicaCount, and triggers.

Breaking down scale-to-zero: when scaling replicas to zero is safe and beneficial, when it's dangerous due to cold start, and how to use activationThreshold to prevent flapping.

Securing scaler access to event sources: secretTargetRef, env, and podIdentity patterns for AWS, Azure, and GCP, plus the difference between TriggerAuthentication and ClusterTriggerAuthentication.

Getting to know four core KEDA scalers: CPU/memory for resource metrics, cron for schedule-based predictive scaling, Prometheus for custom metric queries, and an HTTP scaler based on pending requests. All with ready-to-use ScaledObject examples.

Exploring message queue scalers from cloud and self-hosted sources: AWS SQS with IRSA, Azure Service Bus, GCP Pub/Sub, RabbitMQ, Kafka consumer lag, Redis Streams, and NATS JetStream. Complete with configuration and autoscaling best practices.

Getting to know database-based scalers: PostgreSQL and MySQL with queries, Redis list length, and MongoDB. Then an introduction to the KEDA HTTP Add-on and HTTPScaledObject for scaling HTTP workloads to zero with request buffering.
