I always look at the suggested changes. Always. Not infrequently, LLMs want to put new methods in what I consider the wrong place. Not infrequently, the LLM requires pushback and admits it was over engineering something. Finally, if you can't understand how or where something is implemented then when it comes time to fix it you will likely have a hard time finding what to fix. Just my perspective.

Do you ever use the LLM itself to fix the code? Does the overengineering ever create actual problems, or just perceived ones? Just trying to cut to core of the topic, not trying to say you're right or wrong.

I'm not sure most people do hate llm use as users. I am very impressed, well more like stunned at where we have come in the last few year. I use Claude almost every day and I love the productivity gains it affords.

What i will say though is that (for me anyways) working with an llm like Claude can be mentally taxing because of how fast we can move together. I also refuse to vibe code so i am always proofing the work and maintaining a context of what we are working on

Perhaps this increased cognitive load is what others are talking about if they say they dont like using an llm

> What i will say though is that (for me anyways) working with an llm like Claude can be mentally taxing because of how fast we can move together.

"we". Not creepy at all. AI is not a person.

Sure, I get it.

On the other hand it is not like we (as people) never personify our tools. Take boats as the iconic example. Many languages place genders on nouns. I'll also say that it sure looks like a conversation to me when I use Claude the tool.

Perhaps, jsdocs might help here.

I thought most CEOs had come to the conclusion last week that AI was not increasing productivity.

They must be on last week’s model.

For a personal wiki I would use wikijs

I have a nice (nice to look at but not expensive) skeleton watch and I love it. I would never use a "smart" watch myself.

Perhaps you query lacks sufficient context.

As a counter opinion, I find Opus 4.7 to be great. It is my default model. The only thing I don't like is how verbose it while chatting. I have tried without success to make it more concise and would fall back to 4.6 if it was not producing better code for me.

Maybe I am not doing something right... It failed me at a repetitive task that I have been doing since 5 months now. After a long, verbose chat, it finally delivered only to immediately fail next week.

Ya, I can't get it to shut up most of the time!

The article is all over the map. A script run by one person is "real" but implementing a front end is not?

Bad data makes you confidently wrong. No data means you understand that your decisions are made based on your intuition. It is much better to know the risks than to be unaware of them.

To add to this, it's much harder to turn back while confidently wrong than it is when you know you are building on flimsy assumptions.

By the time you turn around on bad data, much more damage can be done.