So in personal projects I still code things myself. I use LLMs for help (like when I was using Stackoverflow or Github issues) but I do type the code myself (sometimes copy/paste). Another reason for doing this is that I do not trust agents CLIs running in my personal computer (I don't trust OpenAI/anthropic/etc), and setup a VM or container is just too much hassle.
Now at work, I do use agents. I don't mind the CLIs running free in my work laptop (I couldn't care less). Besides the company pays for Claude/OpenAI/Gemini, so they allow us to use them. Certainly I feel less happy when doing work with agents. For instance, I landed in a new codebase recently, and I have been using agents to understand and write minor fixes/features... it does feel "fast" but I ultimately don't get to understand the codebase as good as if I were going to do it the old way (reading the codebase myself and fighting with it myself)
Yes often I do, because I feel way more cognitive load spread across tasks, since I can take on more. However when I want to just get an idea out of my head fast, it does make me happy :-)
I don't know if anyone else is experiencing this "mental creep"?
I’m so happy. I can think about what actually matters and tackle hard problems that were otherwise bottle-necked by how fast I could type syntax correctly on the keyboard.
Not all efficiency is good. Sometimes, programming is more art than science. And for the art side, I find that thinking and struggling is useful and satisfying in a human way.
Agents speed things up, but they also take away some of the small aha moments you get while figuring things out yourself. I think it depends on whether you enjoy the process or just the outcome.
the question conflates two things worth separating: enjoyment of problem-solving versus satisfaction of shipping.if most of your craft satisfaction came from debugging a subtle race condition or working out an elegant abstraction, that aha moment is harder to get when the agent gets there first. that's a real loss and worth naming honestly rather than hand-waving away.but if your satisfaction came from seeing something working, from momentum, from having built something a user can actually touch - agents compress the gap between idea and working software in a way that's hard to argue with.where it gets uncomfortable: watching the agent do the intellectually interesting parts while you review and manage QA. that discomfort is useful signal though. it means you were getting satisfaction from implementation work that, in hindsight, could have been delegated. the natural response is to move upstream - to the parts that still require judgment: what to build at all, which edge cases actually matter to real users, what the right abstraction is.for me as a solo founder it's been net positive. the craft satisfaction shifted, it didn't disappear.
I would offload boring stuffs like HTML, CSS to AI agent.
Core problem solving would start from prompting my brain.
I have been in programming for decade now, building my startup now.
AI as boring co-pilot, I would anyday feel happy.
As an older programmer, I’m actually happier using AI to write code. We already know what good engineering looks like, and now we’ve got a new toy we can learn from and have fun with.
So in personal projects I still code things myself. I use LLMs for help (like when I was using Stackoverflow or Github issues) but I do type the code myself (sometimes copy/paste). Another reason for doing this is that I do not trust agents CLIs running in my personal computer (I don't trust OpenAI/anthropic/etc), and setup a VM or container is just too much hassle.
Now at work, I do use agents. I don't mind the CLIs running free in my work laptop (I couldn't care less). Besides the company pays for Claude/OpenAI/Gemini, so they allow us to use them. Certainly I feel less happy when doing work with agents. For instance, I landed in a new codebase recently, and I have been using agents to understand and write minor fixes/features... it does feel "fast" but I ultimately don't get to understand the codebase as good as if I were going to do it the old way (reading the codebase myself and fighting with it myself)
Yes often I do, because I feel way more cognitive load spread across tasks, since I can take on more. However when I want to just get an idea out of my head fast, it does make me happy :-) I don't know if anyone else is experiencing this "mental creep"?
I’m so happy. I can think about what actually matters and tackle hard problems that were otherwise bottle-necked by how fast I could type syntax correctly on the keyboard.
Not all efficiency is good. Sometimes, programming is more art than science. And for the art side, I find that thinking and struggling is useful and satisfying in a human way.
As a hobbyist and indie hacker, love it. As a professional maintainer - it made a dry job dryer, faster.
Agents speed things up, but they also take away some of the small aha moments you get while figuring things out yourself. I think it depends on whether you enjoy the process or just the outcome.
the question conflates two things worth separating: enjoyment of problem-solving versus satisfaction of shipping.if most of your craft satisfaction came from debugging a subtle race condition or working out an elegant abstraction, that aha moment is harder to get when the agent gets there first. that's a real loss and worth naming honestly rather than hand-waving away.but if your satisfaction came from seeing something working, from momentum, from having built something a user can actually touch - agents compress the gap between idea and working software in a way that's hard to argue with.where it gets uncomfortable: watching the agent do the intellectually interesting parts while you review and manage QA. that discomfort is useful signal though. it means you were getting satisfaction from implementation work that, in hindsight, could have been delegated. the natural response is to move upstream - to the parts that still require judgment: what to build at all, which edge cases actually matter to real users, what the right abstraction is.for me as a solo founder it's been net positive. the craft satisfaction shifted, it didn't disappear.
I would offload boring stuffs like HTML, CSS to AI agent. Core problem solving would start from prompting my brain. I have been in programming for decade now, building my startup now. AI as boring co-pilot, I would anyday feel happy.
As an older programmer, I’m actually happier using AI to write code. We already know what good engineering looks like, and now we’ve got a new toy we can learn from and have fun with.
No