When I first heard about the idea of using AI to generate tests, I was curious and excited, like many of you. At Trendyol iOS team, we have already adapted ways to use Generators to generate repetitive codes for our unit and UI tests. This has greatly improved our ability to write better and more tests. But every good thing comes with a price.
The first issue with generated tests is that they lose the context of checking the flow step by step and making sure everything is developed correctly. Having this superpower can easily lead to generating unit tests for wrong flows. Additionally, people may not review machine-generated code as thoroughly as they would human-generated code, leading to a decrease in the overall quality of the code.
The biggest benefit of having unit tests is that they can help identify errors in future iterations of the code. However, if one has the ability to generate new tests quickly and easily, there may be less incentive to check for errors, and if they always create new tests instead of understanding why they fail, you lose the main purpose. It is important to consider the long-term effects of relying heavily on generated tests and to balance the use of generated tests with manual touches to ensure the overall quality of the code.
Overall, I am excited about the potential of using AI in software testing, and I look forward to continuing to explore this area. I believe that it can be very useful in generating tests for legacy code, or that AI-generated code can be used as a starting point for developers. This can help us achieve a more efficient and effective testing process.
Do you think AI will increase overall software quality or not? Please let me know via Twitter, Linkedin, or email.
Now, onto the this week’s episode.
Reading something you already know well may not teach you much new information, but it serves as a reminder and helps to reinforce your understanding.
🔗 https://medium.com/10-minutes-qa-story/test-design-techniques-cheetsheet-26546c4271c2
Meta has converted its Android codebase to Kotlin, although the migration process did come with some challenges.
🔗 https://engineering.fb.com/2022/10/24/android/android-java-kotlin-migration/
Reddit has become a mobile-first company in order to scale and improve the user experience for its mobile app users. In this recap, they described the in-depth process of adapting to a mobile-first focus and migration process.
🔗https://www.reddit.com/r/RedditEng/comments/zkap1u/reddit_recap_state_of_mobile_platforms_edition/
No posts
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.