Nikos Papaioannou

I have been using Claude code and cursor for the past almost 2 years. Everybody is talking about loop engineering and graphs. I have such a setup with over 10 agents. It used to work pretty well. Over the past month, the more autonomy I give to the agents the more complicated they make the solution thus getting trapped in a loop for hours. They tend to find bugs and add nice to have features that are not even nice to have when the effort is huge and the value is low. Am I the only one experiencing this strange behavior?

I caught myself relying on AI even for the simplest tasks. That s why I decided to downgrade my subscription and make the minimal use of it. At the end of the day, I get this sensation of fulfillment. The quality of work has been increased. However, the speed has been reduced. The issue is that we are reaching a point where we re building stuff because the AI suggested so and not because we actually need it. We are putting less thought into designing and prioritizing and more effort into the nice to have features. Anybody else worrying that the critical thinking skills are in decline due to extensive use of AI? How do you plan to entangle it? Mention personal examples if applicable.