What is the difference between DevOps and Agile? Why data scientists need to understand these concep
During my hunt for a job, I scanned through hundreds of advertisements for data scientists and one of the requirements always appears in the description is understanding of Agile methodology (some other companies require DevOps). Though I have been involved in projects which follow agile methodology, I do not know what is the difference between Agile and DevOps and why a data scientist needs to have this certain knowledge. Thus, this article aims to shed some light on these matters.
What is Agile? Agile Methodology involves continuous iteration of development and testing in the Software Development Life Cycle process. This software development method emphasizes on iterative, incremental, and evolutionary development. You may also have heard of Scrum or Kanban, they are also concepts in agile.
What is DevOps? DevOps is a software development method which focuses on communication, integration, and collaboration among IT professionals to enables rapid deployment of products.
Why do data scientist need to understand Agile/DevOps workflow? Well, in short, this is a quote that I copy from here (if you want to read for more insights):
"Data science is part of software engineering, part of research and innovation, and fully about using data to create impact and value. Aspects of data science that work well with agile tend to be more of the engineering nature, while those closely related to research tends not to fit as well"
There is another article explain the latter idea why some aspects of Data Science does not fit in Agile.
Stakeholders and communication chain in a typical IT process (Source: Guru99)
Agile addresses gaps in Customer and Developer communications (Source: Guru99)
DevOps addresses gaps in Developer and IT Operations communications (Source: Guru99)
Summary of all key differences between Agile and DevOps:
References:
https://www.guru99.com/agile-vs-devops.html
https://towardsdatascience.com/agiledatascience-3b7ca65278a4
https://www.linkedin.com/pulse/why-data-science-doesnt-respond-well-agile-jeffrey-humpherys/