Course · Training · Workshop
Dev Containers: Reproducible Development Environments
One day hands-on: define your development environment as code in devcontainer.json and run it the same way in your local IDE, in GitHub Codespaces, for coding agents and in CI.
A new team member, a second laptop, a coding agent and the CI pipeline all need the same thing: the toolchain of your project in the right versions. A Dev Container describes this environment as code. The file devcontainer.json lives in the repository and defines the container image, the tools, the IDE extensions and the settings. The Development Containers specification is open, and VS Code, IntelliJ IDEA, the Dev Container CLI and GitHub Codespaces work with it. In this course you build a Dev Container for a project in the stack your group chooses: Go, Python, Java, C/C++ or .NET. You start from a template, add tools with Dev Container Features, install IDE extensions and settings through the configuration and run setup steps with postCreateCommand. You then move on to a custom image with a Dockerfile and to a multi-container setup with Docker Compose or Podman Compose, for example with a database next to your application. Coding agents work in the same container. The commands an agent runs execute inside the container, with the tools and dependencies your team uses, and the container serves as a sandbox for the agent. You use GitHub Copilot in agent mode inside the Dev Container, connect a Model Context Protocol (MCP) server and run a small language model with Ollama through a Dev Container Feature. At the end you open the same environment in GitHub Codespaces and reuse it in a GitHub Actions pipeline with the devcontainers/ci action. Laptop, cloud IDE, agent and CI then build and test with one definition from the repository.
What participants say






These customers booked courses in the same topic cluster.More customers →
Content
This training is for software engineers, tech leads and platform engineers who want one development environment for the whole team. You define it once in the repository and use it in the local IDE, in the cloud IDE, for coding agents and in CI.
The course consists of the following topics and can be adapted to the audience. The group chooses the stack for the examples: Go, Python, Java, C/C++ or .NET. For in-house courses we agree on the optional modules, for example the embedded module with C/C++ and ROS 2.
Setup and first Dev Container
- VS Code, the Dev Containers extension and Docker Desktop or Podman
- Open an existing project with “Reopen in Container”
- How the IDE, the container engine and devcontainer.json work together
- Hands-On: start a sample project in a Dev Container and run it
Dev Container basics
- devcontainer.json: image, name and container settings
- Start a configuration from a Dev Container Template
- Add tools with Dev Container Features
- IDE extensions and settings as part of the configuration
- Setup steps after creation with postCreateCommand
- Build the environment with the Dev Container CLI
- Hands-On: build a Dev Container for a project in your stack, from the template to passing tests
AI and agents in Dev Containers
- Coding agents in the container: GitHub Copilot in agent mode
- The container as sandbox for agents: tokens, mounts and network access
- Model Context Protocol (MCP) servers in the project configuration
- Small language models (SLMs) with the Ollama Feature or an external Ollama instance
- Hands-On: connect an MCP server for current documentation and let the agent solve a task in the Dev Container
Custom images and multiple containers
- Your own Dockerfile with stack-specific tools
- Multi-container environments with Docker Compose or Podman Compose, for example with a database
- Containers inside the Dev Container: Docker-in-Docker and Docker-outside-of-Docker
- Hands-On: run a Dev Container with a custom image and a Compose setup with a second service
Cloud development environments
- Open the Dev Container in GitHub Codespaces
- Codespaces settings in devcontainer.json
- Shorter start times with prebuilds and cached images
- Overview of further cloud development environments
- Hands-On: push the project and build and test it in a codespace
Dev Containers in CI
- Run build and tests in the Dev Container with the devcontainers/ci action in GitHub Actions
- Cache the Dev Container image in a container registry
- Hands-On: add a workflow to the project and compare the run times with and without the cached image
Optional modules, depending on audience and time
- Integration testing with Testcontainers in a Dev Container
- Nix in Dev Containers
- Building custom cloud development environments at scale
- Embedded: C/C++ tooling with CMake and GoogleTest, ROS 2 and Gazebo in a Dev Container, connecting hardware to a container
At the end of the day you have a repository with a working Dev Container for your stack, including an MCP configuration and a CI workflow. You can open it locally and in GitHub Codespaces and you know the steps to introduce Dev Containers in your team’s projects.
The actual course content may differ from the above depending on the trainer, delivery, duration and the composition of participants.
Request this course in-house
Request a public date
No suitable public date? Register without obligation — once there is enough interest we schedule a new public date and let you know first.
More about Dev Containers
A development container lets you use a container as a full-featured development environment. The Development Containers specification defines the format of devcontainer.json. Dev Container Features are self-contained, shareable units of installation code and configuration that add tools to an image. Dev Container Templates package the configuration for a complete development environment as a starting point.
The course follows one environment through its stations: from the local IDE to a coding agent, to a cloud development environment and to the CI pipeline. The course builds on the workshop Introduction to Dev Containers by Oleg Nenashev. He presented the embedded variant with a ROS 2 robotics project at the QA&Test conference in 2026 and publishes its material openly.
Further resources:
- Development Containers: specification, Features and Templates
- VS Code: Developing inside a Container
- GitHub Codespaces documentation
- devcontainers/ci: Dev Containers in GitHub Actions
- Claude Code in a Dev Container
- Oleg Nenashev: Introduction to Dev Containers
- Workshop material: Container Developer Environments for embedded projects
Related courses
- GitHub CI/CD for Enterprise: pipelines with GitHub Actions in depth.
- Container & Kubernetes DevOps Introduction: container basics, images and Kubernetes.
- Agentic (AI-Assisted) Software Engineering: working with coding agents in software projects.
- Embedded DevOps: CI/CD for Firmware & Embedded Linux: build, test and release pipelines for embedded software.
