Ask HN: Non AI Contributor on Project Teams
2 comments
Task tracking is now obsolete. Only the slowest most inefficient firms do this. So yes, I am learning as well how tasks are now effectively harder to create than actually implement, I tend to be a bit verbose and sometimes research the codebase for how a feature ought to be executed. All in all my tasks could eventually be done faster by llm’s
Personally, since I only do cosmetic changes to PRs like refactoring them to fit into the current codebase and sprinkling some best practices here and there - I haven’t noticed any decrease in coding velocity.
Personally, since I only do cosmetic changes to PRs like refactoring them to fit into the current codebase and sprinkling some best practices here and there - I haven’t noticed any decrease in coding velocity.
In addition to your experience I notice that through AI contributions our team reaches sprint goals way too early and PMs aren’t catching up with creating new tasks.
Did you notice a significant decrease in delivery speed (including debugging, fixing stuff in AI contributions)?
I wonder whether reducing AI usage might even help with cognitive load. The last few months I felt increasingly exhausted after pushing through PR after PR (although it is quite fun to do this on personal projects).
Did you notice a significant decrease in delivery speed (including debugging, fixing stuff in AI contributions)?
I wonder whether reducing AI usage might even help with cognitive load. The last few months I felt increasingly exhausted after pushing through PR after PR (although it is quite fun to do this on personal projects).
So many features felt flaky. As the team lead who reviews everyones PR and was also in charge of the deployment and monitoring, it created a lot of work for me because I had to fix every hotbug - a case in point was a pagination not working on some page or other.
During some of these fixes, I felt my knowledge of the codebase wasn't as it ought to be especially since I didn't participate as much.
Also, there were a lot of code that was so much more complex that I felt it will be if a human had written it - I felt my review was lacking in some capacity as well.
For a while I struggled reviewing PRs that turned out to be miniature ai slops. And of course, under pressure of meeting deadlines and other things you end up merging PRs that just worked via click-testing without actually reviewing the code.
In the end, I felt the codebased lacked a team member that truly understood all that was going on the way it would've been if we all coded by hand. Since I was the least contributing member of actual code, I decided I will no longer use any coding tools on the codebase - not even autocomplete from copilot.
Since that decision, I've felt my knowledge of the codebase increase and I see the codebase from an entirely different perspective than other developers working on the project. I feel like a non-ai contributor could indeed become an essential member of every development team until agents completely take over to some extent.
Do some of you have a similar experience?
FYI - I use coding tools in my personal projects. But this work project had enough time for us to deliver that AI usage was probably not an economic necessity.