I've felt this shift also. I used to get one method at a time from AI and incorporate it into my code. Now I can give it the whole task and have it write it's own unit tests. The thing is, when it does eventually make a mistake it's harder to find because I don't really know all the code
2026-05-24 16:12:40
1
Agentic Engineering :
👋🤖👋🤖
2026-05-24 15:22:32
0
To see more videos from user @agenticengineering, please go to the Tikwm
homepage.