LLMs _can_ be used as skill multipliers but it’s up to YOU whether you choose to use them as a leverage for your own capabilities or whether you deskill yourself into a helpless slop slinger. 2/2
I fear we’re watching multiple generations to give up any kind of ambition to improve at whatever craft can be automated mediocrely by LLMs. Repeatedly pulling the prompt lever will never end in excellence. ESPECIALLY if the boosters are right, “prompting” is not a long-term valuable skill. 1/2
"The question is: why did I stop auditing when AI became the author?" Why would I let Claude merge a PR that I'd block immediately if a junior dev submitted it? AI didn’t invent bad patterns, but AI scaled them from occasional mistakes to systematic defaults; it amplified all our worst habits.
"Friday, I tried the experiment: let Claude build our component library from our team’s Figma file. Monday, I ran Lighthouse." Before: 187KB, LCP 1.2s, TBT 120ms. After: 524KB, LCP 3.4s, TBT 410ms. https://www.designsystemscollective.com/ai-built-it-in-1-hour-lighthouse-gave-me-43-6f88e5c23d6b
Отличный способ убрать надоедливую анимацию при переключении рабочих столов https://arhan.sh/blog/native-instant-space-switching-on-macos/ #macos #swift
"There will be plenty of work fixing the things that we broke thanks to the most dangerous AI psychosis of all – the hallucinatory belief that "writing code" is the same thing as "software engineering." https://pluralistic.net/2026/01/06/1000x-liability/#graceful-failure-modes?utm_source=tldrdev...
ty is awesome ✨ I'm amazed at Astral's technical excellence and developer experience focus. Ty is not only crazy fast, it improves what's possible with Python (check their Type System docs). And the editor support, chef kiss 🤩 docs.astral.sh/ty/ ty @crmarsh.com
ty
ty is an extremely fast Python type checker.
docs.astral.sh
I know I've glazed Overmind in my last video, but I think I've come to like mprocs more now! It separates the output, is actively maintained, and it works better overall for me: github.com/pvolok/mprocs
GitHub - pvolok/mprocs: Run multiple commands in parallel
Run multiple commands in parallel. Contribute to pvolok/mprocs development by creating an account on GitHub.
github.com
"Tools can't do the learning for you. An AI can generate a perfect solution in seconds, but it cannot give you the experience you gain from the struggle of creating it yourself." https://martinfowler.com/articles/llm-learning-loop.html #llm #learning #programming #llmdevimpact
"In order to be an AI optimist, you have to believe yours is not one of the jobs that risk being automated or downsized, and that you are not among the potentially millions of workers staring down displacement." https://joshcollinsworth.com/blog/sloptimism #llm
"The engineers who understand that programming isn’t the job, the ones who bring judgment, context, and the ability to figure out what to build, those are the ones who stay." https://terriblesoftware.org/2025/12/11/ai-can-write-your-code-it-cant-do-your-job/ #llm #softwarecraft
Рассказываю о своей утилите gh-commitmsg – расширении для GitHub CLI, которое автоматизирует написание commit-сообщений. amgold.ru/blog/gh-comm... #Programming #GitHub #GoLang #CLI
Автоматизированная генерации сообщений conventional commit | Посты | Александр Голдовский
Рассказываю о своей утилите gh-commitmsg – расширении для GitHub CLI, которое автоматизирует написание commit-сообщений.
amgold.ru
Лучше и не скажешь! https://justuse.org/curl/ #devtools
Заслуженная награда🎖️! Рад за Дэниэла. Просто юзайте curl! ✊ https://daniel.haxx.se/blog/2025/10/25/a-gold-ceremony-to-remember/ #devtools
Классная аналогия формирования команды разработчиков https://newsletter.manager.dev/p/build-your-engineering-team-like #leaddev #management
Мигель тестирует производительность нового релиза Питона https://blog.miguelgrinberg.com/post/python-3-14-is-here-how-fast-is-it #python #programming
Саймон предлагает новый термин для обозначения ответственной разработки с LLM. В посте приведены очень правильные практики. https://simonwillison.net/2025/Oct/7/vibe-engineering/#atom-everything #llm #programming
"Software development is a unique craft, it’s equal parts creative and equal parts engineering, two opposing forces that when combined can create something truly amazing. Fight the temptation to add marketing into the mix and dilute the craft with it." https://dayvster.com/blog/dev-culture-is-dy...
Шон реализовал гениальную в своей простоте идею "бесконечной энциклопедии" на основе LLM. Впечатляет! https://seangoedecke.com/endless-wiki/ #llm #inspiration
"Working on projects that never ship is, anecdotally, one of the largest causes of burnout." Мужик дело пишет! https://read.engineerscodex.com/p/how-to-burnout-a-software-engineer #programming
Stop making your job your personality. Pay your bills. Turn off your phone. Leave the spreadsheet in the cloud and go make dinner. You are not a commodity. You are not your engagement metrics. You are not your title or your side hustle or your carefully sharpened personal brand. You are a person....
"Abundance is a moral trial greater than scarcity. Scarcity disciplines. It forces you to structure life around what must be done. Abundance tempts you to dissolve into sloth, grievance, resentment. The great danger of our era is not deprivation. It’s decadence." https://www.joanwestenberg.com/p...
Сделал небольшое расширение для Raycast – справочник по командам NeoVim (LazyVim). Может быть полезно тем, кто как я постоянно забывает нужные команды 😃 github.com/hazadus/rayc... #raycast #neovim #lazyvim #programming
Очень творческий мод для старой игрушки, браво! https://simonwillison.net/2025/Sep/10/animal-crossing-llm/#atom-everything #llm #game
Красивая аналогия с телепортацией. Есть над чем задуматься! https://www.geoffreylitt.com/2025/09/10/ai-as-teleportation #llm