I wrote an essay about why I'm done using AI for coding, even if it costs me my job and career: https://brettcodes.com/im-done-using-ai/

[anderseknert@swecyb.com]

@brettcodes good call! You’re not alone. I’m also coding by hand like a caveman, and I wouldn’t have it any other way

[kcjpop@toot.cafe]

@brettcodes Thanks for writing it. Truly resonates with me. I’m doing the same thing now, going back to learn basic stuff to find joy in coding. Also thanks for making usagi

[happyborg@fosstodon.org]

@brettcodes

[happyborg@fosstodon.org]

@brettcodes good to read this which covers many of the issues I have with #GenAI. It remains frustrating to see so many developers not realising what it's doing to them and society, where this is leading and how absolutely wrong headed the investment in this is compared to other uses of that money.

What you describe as your love of programming is very like my own. I wonder if this is one of the differences. Perhaps if you didn't come to code because of the joy, maybe that's the difference? #LLM

[cesar_@infosec.exchange]

@brettcodes
AI ushered in an era of the dehumanization of technology, especially computing. Along with a pandemic of stupidity. I don't want to use AI for programming either.

[davecb@hachyderm.io]

@brettcodes
I du use LLMs, but not to code. I use like lint — a fallible tool that suggests mistakes I made, ones I evaluate for myself.

I stole the entire idea from Advait Sarkar. It was in the October 2024 Communications of the Association For Computing Machinery.

https://cacm.acm.org/opinion/ai-should-challenge-not-obey/

where he wrote "AI Should Challenge, Not Obey: Let’s transform our robot secretaries into Socratic gadflies."

I wrote a multi-part rant on not using to write/codet: this is from part 3. https://leaflessca.wordpress.com/2026/06/07/hedgewitch-part-3-llms-should-challenge-not-obey/

[elhult@mastodon.acc.sunet.se]

@brettcodes This morning, I was thinking if this happens in research as well. It makes me arguably faster. But does it make me worse?

With suitable MCP servers, it helps me proof read manuscripts, cross ref vs my lab diary, inspect my citation management software etc.

But do I deskill myself like this?

[toerror@mastodon.gamedev.place]

@brettcodes The comment regarding 3x more productive not resulting in higher turnover / etc; I think what exacerbates this problem is the widely held view in industry that developers are builders and if they aren't building then they represent an ongoing sunk cost, so you always ended up with devs being assigned development busy work... The reality in most small to medium orgs is that the developers become the product custodians, despite what the business thinks is going on.

[weebull@mastodon.social]

@brettcodes I think your observation that "this isn't the job you agreed to" is key. Let the people who are ok driving themselves into the ground chasing "productivity" do so.

You can go be a craftsman and enjoy what you do.

Here's to a second luddite uprising. Let's hope it doesn't end the same way.

[javascript@app.wafrn.net]

@brettcodes@mastodon.social

I wish we could all find such a place. A place to work where people cared for their creations. A place where creating the stuff that makes people interface with computers was about people, not just about "number go up"

[andnull@social.nouveau.community]

@brettcodes This is a nice change of pace from all the people saying they "can't deny the utility of AI" while also saying "they are uncomfortable with the harms". Hopefully you can keep your job (and maybe find others to join your side of things).

My non-tech job has a very easy contradiction w/r/t AI systems so I have been providing that explination to all my coworkers whenever I hear them express confusion as to why the system wasn't usedful.

[andnull@social.nouveau.community]

@brettcodes (granted AI systems are nothing but contradiction. It's just in this case I can more or less prove it doubles our work load by being the exact same shape of the job we already do)

[etch9@tech.lgbt]

@brettcodes
a great first post for the fediverse ahahah

[daseinslaborant@berlin.social]

@brettcodes Very inspiring read, and what a great approach to go back to where the fun is in creating software, and in the end also invest into the relationship between you and your team.

[jakubzet@mastodon.social]

@brettcodes It's very conforming to read someone thinking similar, having the same feelings around AI-transformed work (and life) - and even more, to see such a positive reaction around the YT and Fedi! Wishing you, myself, and other devs around the globe the very best! ️

[brettcodes@mastodon.social]

@Etch9 I didn't expect such a response lol! I don't really use social media, so I'm pleasantly surprised it's resonating with people.

[paco@infosec.exchange]

@brettcodes You’re in good company here. Welcome aboard.

[splendorr@mastodon.social]

@brettcodes Thank you so much for sharing details of this whole journey. I agreed with your observations, and have seen others go through this process in their own ways. I’m really glad you could recognize the effects on so many aspects of your wellbeing, and decide to stop!

[splendorr@mastodon.social]

@elhult @brettcodes Short answer is: yes! You’re absolutely diminishing your own understanding, and undermining the validity of your research.

Making real knowledge takes time. It just does! It’s a feature of the process, and the discoveries and connections you might make along the way can only happen in that time-based process.

Good luck!

[shimst3r@famichiki.jp]

@brettcodes observing my coding colleagues from next door, I see how they slowly lose their spark, their eye for detail, and their will to discuss or refine requirements.

"Why do it right if you can do it fast and adapt?"

this "move fast and break things" killed everything