Skip to content

Intro to DevOps

What is DevOps?

DevOps is a methodological framework that integrates software development and IT operations. By leveraging automation, collaborative cultures, and consistent environments, DevOps enables teams to build, test, and deploy high-quality software with increased velocity and reliability.

Typical DevOps tools include:

  • Version control systems (e.g., Git)
  • Collaboratino, continuous integration/continuous deployment (CI/CD) platforms (e.g., GitHub, GitLab)
  • Containerization tools (e.g., Docker)
  • Container orchestration platforms (e.g., Kubernetes)

Why we need DevOps?

Version control keeps code history clean

Instead of copying files like analysis_final_v2.py or commenting out old code, we can track changes properly with Git and return to earlier versions when needed. For example:

  • Messy commented out old code lines: 

    def get_squares(numbers):
    # old version
    # result = []
    # for n in numbers:
    #     result.append(n ** 2)
    # return result

    # new version
    return [n ** 2 for n in numbers]

  • With Git, we keep only the latest version, but still preserve the previous one through commit history: 

    def get_squares(numbers):
    return [n ** 2 for n in numbers]
    Commit history

  • Likewise, no need to create multiple files or folders to track code versions. Commit history tracks this. Different version history

Remote repository and collaboration

Remote deployment

  • Deploy your website to GitHub Pages using GitHub Action. Example

  • Deploy your ML application to Hugging Face Spaces. hf space

Testing and CI/CD catch problems early

Automated checks help us verify that code still works before it is merged or deployed, which is much safer than testing everything manually at the end.

Machine-agnostic development avoids "works on my machine" issues

With consistent environments such as containers, the same code can run more predictably across laptops, servers, and cloud machines.

Machine-agnostic deployment improves reproducibility

If development and deployment environments are consistent, it becomes much easier to move a project from local testing to remote execution without unexpected failures.