Who this programme is for
Graduates and working engineers who want to own how software is built, shipped and run. Some programming exposure helps; comfort on a Linux command line is non-negotiable and we build it in the first module.
What you will be able to do
- Operate confidently on Linux and automate it with shell and Python
- Containerise an application and run it on Kubernetes without copying manifests blindly
- Build a CI/CD pipeline from commit to production with real gates
- Provision cloud infrastructure as code with Terraform
- Instrument a service and debug an incident from metrics and logs
Syllabus
Module 1 — Linux and networking
Filesystem and permissions, processes and systemd, package management, shell scripting, SSH, DNS, TLS, and the networking model you need before anything else makes sense.
Module 2 — Git and collaboration
Branching strategies, rebase versus merge, pull request workflow, resolving the conflicts that scare people, and repository hygiene.
Module 3 — Containers
Docker images and layers, writing a Dockerfile that is small and reproducible, multi-stage builds, registries, Compose for local stacks, and container security basics.
Module 4 — Kubernetes
Pods, deployments and services, ingress, config maps and secrets, probes, resource requests and limits, autoscaling, RBAC, Helm, and debugging a cluster that is misbehaving.
Module 5 — Infrastructure as code
Terraform language and state, modules, planning and applying safely, drift, and configuration management with Ansible.
Module 6 — CI/CD
Pipeline design, build and test stages, artefact management, deployment strategies including blue-green and canary, secrets handling, and rollback that works under pressure.
Module 7 — Observability and reliability
Metrics with Prometheus, dashboards with Grafana, centralised logging, alerting that does not cry wolf, SLOs and error budgets, and incident response.
Module 8 — Capstone
Take an application from source to a monitored production deployment: infrastructure as code, a container build, a Kubernetes rollout, a full pipeline and an observability stack — then run a deliberate incident and recover from it.