Most of what makes a productive DevOps engineer is not hidden inside vendor portals - it lives in open source, on GitHub, and it is free. The projects below are the ones I return to most often, whether for learning, daily tooling, or reference implementations of patterns that would otherwise take weeks to work out alone.

DevOps and Site Reliability Engineering (SRE)

Resources to calibrate what good looks like in the discipline.

  • Awesome Scalability - a reading list illustrating patterns of scalable, reliable, performant large-scale systems
  • Awesome Site Reliability Engineering - a curated list of SRE and production engineering resources
  • DevOps Exercises - interview-style questions and practical exercises, useful whether or not you are job-hunting
  • DevOps Resources - companion to the above, with a structured learning roadmap
  • DevOps Roadmap - a step-by-step visual guide to becoming a DevOps engineer, updated yearly
  • roadmap.sh DevOps - the DevOps track from the widely-used roadmap.sh project
  • DevOps Tools - GitHub’s own curated collection
  • How They SRE - publicly available SRE resources organised by company, invaluable for benchmarking practices

Kubernetes

Day-to-day operational tooling that pays back the installation time within a week.

  • k9s - a terminal UI for interacting with Kubernetes clusters; once you use it, the raw kubectl workflow feels archaic
  • kubectx / kubens - fast context and namespace switching, a small tool that saves real time every day
  • Lens - desktop IDE for Kubernetes, good for visual inspection and debugging
  • kOps - provisioning, upgrading, and lifecycle management for production-grade Kubernetes clusters
  • KoPylot - AI-assisted Kubernetes operations, an interesting pointer to where cluster management is heading
  • kube-state-metrics - surfaces Kubernetes object state as Prometheus metrics
  • Kubernetes Dashboard - the official web UI
  • Kubespray - cluster lifecycle management with Ansible playbooks
  • minikube - local Kubernetes clusters on macOS, Linux, and Windows
  • kind - Kubernetes-in-Docker, faster than minikube for CI use cases
  • Popeye - sanitises running clusters, flags misconfigurations and anti-patterns
  • k3s - lightweight Kubernetes distribution, excellent for edge and home-lab use

Linux and the terminal

Modern replacements for the classic Unix toolchain. Adopting even two or three of these will change how you work.

  • bat - cat with syntax highlighting and Git-aware diff marks
  • ripgrep - the grep replacement, fast enough to change how you search code
  • fd - a simple, fast alternative to find
  • fzf - fuzzy finder, pairs beautifully with shell history and file navigation
  • pueue - command-line task queue for managing long-running jobs
  • specctl - translates Kubernetes objects to ECS and vice versa, useful when migrating between orchestrators
  • Test your SysAdmin skills - interview questions for systems administrators

Workflow and Scheduling

  • Airflow - the long-established Python-based workflow orchestrator, still the default for data pipelines
  • Prefect - a modern alternative to Airflow with a more Pythonic API
  • Dagster - data-aware orchestration, strong type system, popular with data-engineering teams
  • dkron - distributed, fault-tolerant cron for cloud-native environments
  • Temporal - workflow engine for long-running, stateful applications, used heavily in fintech

Infrastructure as Code and GitOps

  • Terraform - the de-facto standard for declarative infrastructure
  • OpenTofu - the community fork of Terraform, worth watching as the licensing situation evolves
  • Pulumi - infrastructure as code in real programming languages (TypeScript, Python, Go)
  • Argo CD - GitOps continuous delivery for Kubernetes
  • Flux - the other major GitOps controller