If I would start DevOps from 0 - How would I start and what would I learn

This video presents a pragmatic, project-driven DevOps roadmap, arguing that learners should first master foundational skills like Linux, Bash, and Git before moving to core cloud concepts and a handful of key AWS/Azure services. It advises against trying to become an expert in everything, recommending instead that roughly 20% of Docker/Kubernetes covers most use cases and that learners pick just one CI/CD tool to build end-to-end pipelines from commit to deployment. The speaker emphasizes that treating CI/CD as the core of DevOps and dedicating one to two months to it is essential, while observability tools like Prometheus and Grafana only require a basic working knowledge. A major pitfall highlighted is learning tools in isolation without combining them, as shallow courses leave learners with hidden gaps that make them feel unqualified in interviews. The right approach, the video argues, is to continuously apply and combine new skills on real projects, using your existing background as a foundation. Ultimately, the video promotes a DevOps Bootcamp that bridges isolated tool learning and real-world understanding by teaching how technologies connect.

• ▶ 0:17 Don't skip the foundation phase—mastering Linux, Bash, and Git first makes every later DevOps skill much easier to learn and stack.
• ▶ 1:25 In the cloud phase, learn core cloud concepts first (compute, storage, networking) rather than chasing certifications; master only about five AWS/Azure services instead of getting overwhelmed.
• ▶ 4:00 For containers and CI/CD, you don't need to be an expert—roughly 20% of Docker/Kubernetes covers 80% of use cases, and you only need to pick one CI/CD tool (Jenkins, GitHub Actions, or GitLab).
• ▶ 4:52 Build actual end-to-end pipelines from code commit to Kubernetes/AWS, rather than just reading docs or watching tutorials.
• ▶ 5:12 Treat CI/CD as the core of DevOps and dedicate 1–2 months to mastering it.
• ▶ 5:25 For observability, learn the fundamentals of Prometheus/Grafana in about one month—enough to implement it, not to become an expert.
• ▶ 6:12 Learning tools in isolation (e.g., separate AWS, Docker, and Terraform courses) without combining them is the biggest mistake; real DevOps work requires building workflows that connect tools.
• ▶ 6:32 Shallow courses leave many missing pieces that learners don't realize they lack, and completing them alone doesn't make you valuable as an engineer.
• ▶ 7:01 The right approach is to continuously apply and combine what you learn with what you already know—building iteratively on real projects instead of just studying theory.
• ▶ 7:36 Most learners recognize their knowledge gaps through insecurity, like feeling unqualified from job descriptions or struggling in interviews with topics they never truly learned.
• ▶ 7:59 The DevOps Bootcamp focuses on bridging isolated tool learning and real-world understanding by teaching underlying concepts and how technologies connect and work together.
• ▶ 8:22 The roadmap can be personalized using your existing background—such as software development, QA, sysadmin, network, or database skills—so you build DevOps knowledge on a foundation you already have.

• ▶ 0:00 The Five-Phase DevOps Roadmap (0:00 - 4:49) - - Covers the introduction and details all five phases, from foundations to CI/CD.
• ▶ 4:49 Building Real Pipelines and Observability (4:49 - 6:12) - - Focuses on hands-on CI/CD projects, Prometheus/Grafana monitoring, and why a few months of focused learning is enough.
• ▶ 6:12 Avoiding Isolated Learning and Applying Continuously (6:12 - 7:36) - - Explains the biggest mistake of learning tools in isolation and the right way to combine and apply knowledge.
• ▶ 7:36 Bridging Gaps and Personalizing the Roadmap (7:36 - 9:41) - - Discusses knowledge gaps, the DevOps Bootcamp, tailoring the roadmap to your background, and final next steps.