Quick spin with Amazon CodeGuru Support for Python
After the announcement at AWS Re:Invent 2020 and release of Amazon CodeGuru support for Python, I decided to give it a quick spin and share my findings below.
What is Amazon CodeGuru?
According to the official website:
Amazon CodeGuru is a developer tool that provides intelligent recommendations to improve your code quality and identify an application’s most expensive lines of code.Integrate CodeGuru into your existing software development workflow to automate code reviews during application development, continuously monitor application performance in production, provide recommendations and visual clues for improving code quality and application performance, and reduce overall cost.
This service from AWS leverages years of technical knowledge and experience from Amazon coupled with machine learning experience. Initially, Amazon CodeGuru only supported Java, and now supports Python.
Why Python?
I looked at data ranking programming languages in 2020 on the Computer.org website, IEEE Computer Society website, Python ranks as number 1. In addition, the growing need for data experts (machine learning, data science, artificial intelligence) can also be assumed as a leading indicator. With many companies are looking to compete on analytics and AI, hiring topnotch talent to build their AI-powered business outcomes, it is imperative to higher the best — not only in culture-fit but also technical competence.
In a subsequent post, I will demonstrate how to integrate Amazon CodeGuru into recruitment process.
Taking Amazon CodeGuru Python for a Spin
The picture above shows the Amazon CodeGuru web console. I must commend the AWS design team, it sure looks really sleek. Next, I connected Amazon CodeGuru Reviewer to CD4AutoML GitHub repository to check if I wrote some bugs into the application and potential code improvements.
Once the repository was associated, it was time for Amazon CodeGuru Reviewer to work. I created a pull request on the CD4AutoML GitHub project and Amazon CodeGuru Reviewer started automatically. At this point, I want to point out that no follow email nor reminder was required — an opportunity for RPA to improve developer productivity with Amazon CodeGuru.
For the next couple of minutes, I waited anxiously for the recommendations from Amazon CodeGuru Reviewer. Largely because I was curious to see the bugs I introduced [😊 ] while developing the project, and I’d presented it at so many talks.
Lo!
My code checks out fine and no bugs (Probably, I am a good Python programmer). Amazon CodeGuru Reviewer completed this in less than 5 mins (~3 minutes), which is way lesser than what it would take any human code reviewer to review 128 lines of code.
This completes my quick spin with Amazon CodeGuru Reviewer for Python. In a followup post, I would demonstrate how to simplify/automate code challenge reviews when hiring with Amazon CodeGuru Reviewer.
