AI coding

But, what's with the self driving hate? I take Waymos on a regular basis, and he is basing his credibility on the claim that they are not a thing. Makes him sound bitter more than insightful.

I’m could have used an LLM to assist but then I wouldn’t have learned much.

But I did use an LLM to make a management wrapper to present a menu of options (cli right now) and call the scripts. That probably saved me an hour, easily.

That’s my comfort level for anything even remotely “complicated”.

I find this hilarious. From what I've seen watching people do it, it changes the job from deep thought and figuring out a good design to pulling a lever on a slot machine and hoping something good pops out.

The studies that show diminished critical thinking have matched what i saw anecdotally pairing with people who vibe coded. It replaced deep critical thinking with a kind of faith based gambler's mentality ("maybe if i tell it to think really hard it'll do it right next time...").

The only times ive seen a notable productivity improvement is when it was something not novel that didnt particularly matter if what popped out was shit - e.g. a proof of concept, ad hoc app, something that would naturally either work or fail obviously, etc. The buzz people get from these gamblers' highs when it works seems to make them happier than if they didnt use it at all though.

* Getting good results from AI forced me to think through and think clearly - up front and even harder.

* AI almost forces me to structure and break down my thoughts into smaller more manageable chunks - which is a good thing. (You can't just throw a giant project at it - it gets really far off from what you want if you do that.)

* I have to make it a habit of reading what code it has added - so I understand it and point to it some improvements or rarely fixes (Claude)

* Everyone has what they think are uninteresting parts of a project that they have to put effort into to see the bigger project succeed - AI really helps with those mundane, cog in the wheel things - it not only speeds things up, personally it gives me more momentum/energy to work on the parts that I think are important.

* It's really bad at reusability - most humans will automatically know oh I have a function I wrote to do this thing in this project which I can use in that project. At some point they will turn that into a library. With AI that amount of context is a problem. I found that filling in for AI for this is just as much work and I best do that myself upfront before feeding it to AI - then I have a hope of getting it to understand the dependency structure and what does what.

* Domain specific knowledge - I deal with Google Cloud a lot and use Gemini for understanding what features exist in some GCP product and how I can use it to solve a problem - works amazingly well to save me time. At the least optioning the solution is a big part of work it makes easier.

* Your Git habits have to be top notch so you can untangle any mess AI creates - you reach a point where you have iterated over a feature addition using AI and it's a mess and you know it went off the rails after some point. If you just made one or two commits now you have to unwind everything and hope the good parts return or try to get AI to deal with it which can be risky.

Think about this. Before there were cars on roads, people were just as much happy. Cars came, cities were redesigned for cars with buildings miles apart, and commuting miles became the new norm. You can no longer say cars are useless because the context around them has changed to make the cars a basic need.

AI does same thing. It changes the context in which we work. Everyone expects you use AI (and cars). It becomes a basic need, though a forced one.

To go further, hardly anything produced by science or technology is a basic need for humans. The context got twisted, making them basic needs. Tech solutions create the problems which they claim to solve. The problem did not exist before the solution came around. That's core driving force of business.