Build the pipeline, the infrastructure and the on-call that comes with it.
For people moving into DevOps from support, sysadmin or development, and for DevOps engineers moving toward platform and SRE work. Cloud provider and tooling are chosen per cohort.
Who this track is for
- Support and helpdesk engineers who want to move into infrastructure
- System administrators moving from hand-managed servers to cloud and infrastructure as code
- Developers asked to own CI/CD and Kubernetes
- DevOps engineers moving toward platform engineering and SRE
Three entry levels
Profiling places you on one. The cohort is built from people on the same level with the same target, so the plan is never too slow for half the room.
- level 1
Entry
- you are
- You know Linux basics and some scripting, but have not run CI/CD or cloud infrastructure yourself.
- you get
- You build pipelines, write Terraform and run containers in Kubernetes for a real service, the standard way.
- level 2
DevOps engineer
- you are
- You run pipelines and some cloud already, mostly learned on the job, with gaps you can feel.
- you get
- A complete platform: infrastructure as code, GitOps, observability and incident practice, with the gaps closed.
- level 3
Platform / SRE
- you are
- You want to design platforms other teams build on and own reliability for them.
- you get
- SLOs and error budgets, platform APIs, cost control and the leadership side of on-call.
What a typical plan covers
12 modules in the order a typical cohort runs them. Your cohort's plan is cut, extended and re-ordered after profiling, and names the exact tools.
Linux and networking for operators
01- Processes, systemd, filesystems
- TCP/IP, DNS, TLS
- Debugging with the standard tools
Git-driven operations
02- Repositories as the source of truth
- Branching and review for infrastructure
- Change history you can audit
CI/CD pipelines
03- GitHub Actions and GitLab CI
- Artefacts, caching, environments
- Approvals and promotion
Containers and registries
04- Docker images and multi-stage builds
- Scanning and signing
- Tagging strategy
Infrastructure as code
05- Terraform / OpenTofu modules
- State, workspaces, locking
- Policy checks before apply
Configuration management
06- Ansible for machines that are not containers
- Inventories, roles, idempotency
- Secrets in playbooks
Kubernetes
07- Workloads, services, ingress
- Storage, config, secrets
- Helm, resource limits, autoscaling
GitOps
08- Argo CD or Flux
- Environments and promotion
- Detecting and fixing drift
Observability
09- Prometheus, Grafana, Loki
- OpenTelemetry collectors
- Alerts that mean something
Cloud (chosen per cohort)
10- AWS, Azure or GCP
- IAM, networking, managed services
- Cost: reading the bill, cutting it
Incidents and on-call
11- Runbooks and paging
- Post-incident reviews
- SLOs and error budgets
Capstone
12- A platform for a sample product
- From empty cloud account to monitored production
- Handover document and presentation
Tools you will work with
- Linux
- Bash
- Git
- GitHub Actions
- GitLab CI
- Docker
- Terraform / OpenTofu
- Ansible
- Kubernetes
- Helm
- Argo CD
- Prometheus
- Grafana
- Loki
- OpenTelemetry
- AWS / Azure / GCP
The final list is set in the cohort plan. Where the market is split, such as cloud providers, the cohort picks the one most of its members need.
By the end, you can
- Stand up a production-grade environment from code in an afternoon
- Deploy to Kubernetes through a pipeline with review, promotion and rollback
- Run an on-call shift with runbooks, alerts and a post-incident review
- Explain the cost and reliability trade-offs of what you built
Eight people, AWS and Kubernetes, 14 weeks
Three sysadmins, three developers and two support engineers moving into DevOps. AWS was chosen because six of eight had it in their job targets. The mentor team came from AWS and Kubernetes practice, with a guest on-call lead for the incident module.
Illustrative cohort, built from the typical mix for this track.
Tell us where you want to get to.
Applying takes five minutes. Profiling is free and you keep the skills map. The cohort, the plan and the price come before any commitment.
Cohorts are forming now for all four tracks. We reply within two working days.