home.social

#vibecoding — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #vibecoding, aggregated by home.social.

  1. For context:

    We went from “cannot understand the difference between C and PHP” to “can sometimes write a valid function” to “works reasonably well to work on single files” to “can build a full greenfield app but needs extensive guidance on architecture and APIs” to “can build a full app with an engineer in the loop and build on top of it for a few weeks” to “decent at architecture and can build smaller systems without guidance” in 3 years.

    But when I was trying to talk about labor issues and it being a paradigm shift for the industry at large, the standard response was that I was deluded and spreading FUD. The take that the tools are useless has been constant too, except the goal posts constantly move to whatever the current state of the art. Another take that never dies is that using llm based tools somehow can’t involve skill, that there is no difference between the prompting of an experienced software engineer who has spent years working with llms and the 3 prompts one has put into a random model “to try things out”. Imagine someone coming to like Elixir from Java, typing a few classes in Java, runs it and gets errors and say “elixir is kinda useless, all I got to run was this super barebones program after 17 tries and lots of compile errors”.

    Whether one like using these tools or not (especially if you don’t like them), and especially if you are relatively new to them, spend just a few minutes or hours to compare how far you get with llama (the OG) and pure copy paste by hand, to a newer 8B model in an agent harness, to a model like glm5.1 to gpt5.5 or opus4.6 in a harness.

    That’s the last 2 years in a bottle.

  2. For context:

    We went from “cannot understand the difference between C and PHP” to “can sometimes write a valid function” to “works reasonably well to work on single files” to “can build a full greenfield app but needs extensive guidance on architecture and APIs” to “can build a full app with an engineer in the loop and build on top of it for a few weeks” to “decent at architecture and can build smaller systems without guidance” in 3 years.

    But when I was trying to talk about labor issues and it being a paradigm shift for the industry at large, the standard response was that I was deluded and spreading FUD. The take that the tools are useless has been constant too, except the goal posts constantly move to whatever the current state of the art. Another take that never dies is that using llm based tools somehow can’t involve skill, that there is no difference between the prompting of an experienced software engineer who has spent years working with llms and the 3 prompts one has put into a random model “to try things out”. Imagine someone coming to like Elixir from Java, typing a few classes in Java, runs it and gets errors and say “elixir is kinda useless, all I got to run was this super barebones program after 17 tries and lots of compile errors”.

    Whether one like using these tools or not (especially if you don’t like them), and especially if you are relatively new to them, spend just a few minutes or hours to compare how far you get with llama (the OG) and pure copy paste by hand, to a newer 8B model in an agent harness, to a model like glm5.1 to gpt5.5 or opus4.6 in a harness.

    That’s the last 2 years in a bottle.

    #llm #llms #genai #claude #vibecoding

  3. For context:

    We went from “cannot understand the difference between C and PHP” to “can sometimes write a valid function” to “works reasonably well to work on single files” to “can build a full greenfield app but needs extensive guidance on architecture and APIs” to “can build a full app with an engineer in the loop and build on top of it for a few weeks” to “decent at architecture and can build smaller systems without guidance” in 3 years.

    But when I was trying to talk about labor issues and it being a paradigm shift for the industry at large, the standard response was that I was deluded and spreading FUD. The take that the tools are useless has been constant too, except the goal posts constantly move to whatever the current state of the art. Another take that never dies is that using llm based tools somehow can’t involve skill, that there is no difference between the prompting of an experienced software engineer who has spent years working with llms and the 3 prompts one has put into a random model “to try things out”. Imagine someone coming to like Elixir from Java, typing a few classes in Java, runs it and gets errors and say “elixir is kinda useless, all I got to run was this super barebones program after 17 tries and lots of compile errors”.

    Whether one like using these tools or not (especially if you don’t like them), and especially if you are relatively new to them, spend just a few minutes or hours to compare how far you get with llama (the OG) and pure copy paste by hand, to a newer 8B model in an agent harness, to a model like glm5.1 to gpt5.5 or opus4.6 in a harness.

    That’s the last 2 years in a bottle.

    #llm #llms #genai #claude #vibecoding

  4. For context:

    We went from “cannot understand the difference between C and PHP” to “can sometimes write a valid function” to “works reasonably well to work on single files” to “can build a full greenfield app but needs extensive guidance on architecture and APIs” to “can build a full app with an engineer in the loop and build on top of it for a few weeks” to “decent at architecture and can build smaller systems without guidance” in 3 years.

    But when I was trying to talk about labor issues and it being a paradigm shift for the industry at large, the standard response was that I was deluded and spreading FUD. The take that the tools are useless has been constant too, except the goal posts constantly move to whatever the current state of the art. Another take that never dies is that using llm based tools somehow can’t involve skill, that there is no difference between the prompting of an experienced software engineer who has spent years working with llms and the 3 prompts one has put into a random model “to try things out”. Imagine someone coming to like Elixir from Java, typing a few classes in Java, runs it and gets errors and say “elixir is kinda useless, all I got to run was this super barebones program after 17 tries and lots of compile errors”.

    Whether one like using these tools or not (especially if you don’t like them), and especially if you are relatively new to them, spend just a few minutes or hours to compare how far you get with llama (the OG) and pure copy paste by hand, to a newer 8B model in an agent harness, to a model like glm5.1 to gpt5.5 or opus4.6 in a harness.

    That’s the last 2 years in a bottle.

    #llm #llms #genai #claude #vibecoding

  5. For context:

    We went from “cannot understand the difference between C and PHP” to “can sometimes write a valid function” to “works reasonably well to work on single files” to “can build a full greenfield app but needs extensive guidance on architecture and APIs” to “can build a full app with an engineer in the loop and build on top of it for a few weeks” to “decent at architecture and can build smaller systems without guidance” in 3 years.

    But when I was trying to talk about labor issues and it being a paradigm shift for the industry at large, the standard response was that I was deluded and spreading FUD. The take that the tools are useless has been constant too, except the goal posts constantly move to whatever the current state of the art. Another take that never dies is that using llm based tools somehow can’t involve skill, that there is no difference between the prompting of an experienced software engineer who has spent years working with llms and the 3 prompts one has put into a random model “to try things out”. Imagine someone coming to like Elixir from Java, typing a few classes in Java, runs it and gets errors and say “elixir is kinda useless, all I got to run was this super barebones program after 17 tries and lots of compile errors”.

    Whether one like using these tools or not (especially if you don’t like them), and especially if you are relatively new to them, spend just a few minutes or hours to compare how far you get with llama (the OG) and pure copy paste by hand, to a newer 8B model in an agent harness, to a model like glm5.1 to gpt5.5 or opus4.6 in a harness.

    That’s the last 2 years in a bottle.

    #llm #llms #genai #claude #vibecoding

  6. #Spotify says its best developers haven’t written a line of code since December, thanks to AI - February 2026:

    techcrunch.com/2026/02/12/spot

    Also Spotify, May 2026:

    "Hey everyone!

    We've received some reports mentioning that the app, support site and the Web Player are slow or not working properly. This is being investigated.

    Cheers."

    community.spotify.com/t5/Ongoi

    I'm so tired of buggy #software!

    #VibeCoding #LLMs #Slop

  7. #Spotify says its best developers haven’t written a line of code since December, thanks to AI - February 2026:

    techcrunch.com/2026/02/12/spot

    Also Spotify, May 2026:

    "Hey everyone!

    We've received some reports mentioning that the app, support site and the Web Player are slow or not working properly. This is being investigated.

    Cheers."

    community.spotify.com/t5/Ongoi

    I'm so tired of buggy #software!

    #VibeCoding #LLMs #Slop

  8. #Spotify says its best developers haven’t written a line of code since December, thanks to AI - February 2026:

    techcrunch.com/2026/02/12/spot

    Also Spotify, May 2026:

    "Hey everyone!

    We've received some reports mentioning that the app, support site and the Web Player are slow or not working properly. This is being investigated.

    Cheers."

    community.spotify.com/t5/Ongoi

    I'm so tired of buggy #software!

    #VibeCoding #LLMs #Slop

  9. #Spotify says its best developers haven’t written a line of code since December, thanks to AI - February 2026:

    techcrunch.com/2026/02/12/spot

    Also Spotify, May 2026:

    "Hey everyone!

    We've received some reports mentioning that the app, support site and the Web Player are slow or not working properly. This is being investigated.

    Cheers."

    community.spotify.com/t5/Ongoi

    I'm so tired of buggy #software!

    #VibeCoding #LLMs #Slop

  10. #Spotify says its best developers haven’t written a line of code since December, thanks to AI - February 2026:

    techcrunch.com/2026/02/12/spot

    Also Spotify, May 2026:

    "Hey everyone!

    We've received some reports mentioning that the app, support site and the Web Player are slow or not working properly. This is being investigated.

    Cheers."

    community.spotify.com/t5/Ongoi

    I'm so tired of buggy #software!

    #VibeCoding #LLMs #Slop

  11. Waiting on codex reminds me of waiting for computers to start up or launch a program in the 90s.

  12. Amazing Spider-Man #28
    Written by Joe Kelly, art by Cory Smith and Francesco Manna.

    "Worst worst, you vibe coded with A.I., which makes me puke in my mouth"

    #Comics #VibeCoding #SpiderMan

  13. #venting

    I find it remarkable how #Microsoft have managed to turn #Office, their flagship productivity suit into a smelly trashfire.

    I credit the "#AI revolution" and a culture of #VibeCoding. Product #quality has gone thoroughly downhill as result.

    So happy I don't pay for it out of pocket and I wish the organizations I work with would seriously consider transitioning to alternatives.

  14. RE: mastodon.social/@daringfirebal

    To some, this article may seem to start out kinda slow, but after laying some groundwork, it builds and builds until I literally Laughed Out Loud *TWICE* further in!

    ⭐️⭐️⭐️⭐️⭐️

    👍👍

    #LOL #VibeCoding #Mac #Software #Review #KneeSlapper #YMMV

  15. When did developers start implementing searches that just parse every single letter as an individual search? For example, searching Mescaline Maniacs inside my collection on the Bandcamp app pulled up the useless results below. Mescaline Maniacs was at the top of the list, but wtf are these results even for? This isn't the first place I have encountered this either.

    #enshittification #why #MakeEverythingUseless #breakitall #ai #vibecodebaby #vibecoding

  16. When did developers start implementing searches that just parse every single letter as an individual search? For example, searching Mescaline Maniacs inside my collection on the Bandcamp app pulled up the useless results below. Mescaline Maniacs was at the top of the list, but wtf are these results even for? This isn't the first place I have encountered this either.

    #enshittification #why #MakeEverythingUseless #breakitall #ai #vibecodebaby #vibecoding

  17. When did developers start implementing searches that just parse every single letter as an individual search? For example, searching Mescaline Maniacs inside my collection on the Bandcamp app pulled up the useless results below. Mescaline Maniacs was at the top of the list, but wtf are these results even for? This isn't the first place I have encountered this either.

    #enshittification #why #MakeEverythingUseless #breakitall #ai #vibecodebaby #vibecoding

  18. When did developers start implementing searches that just parse every single letter as an individual search? For example, searching Mescaline Maniacs inside my collection on the Bandcamp app pulled up the useless results below. Mescaline Maniacs was at the top of the list, but wtf are these results even for? This isn't the first place I have encountered this either.

    #enshittification #why #MakeEverythingUseless #breakitall #ai #vibecodebaby #vibecoding

  19. When did developers start implementing searches that just parse every single letter as an individual search? For example, searching Mescaline Maniacs inside my collection on the Bandcamp app pulled up the useless results below. Mescaline Maniacs was at the top of the list, but wtf are these results even for? This isn't the first place I have encountered this either.

    #enshittification #why #MakeEverythingUseless #breakitall #ai #vibecodebaby #vibecoding

  20. Audio guestbook ☎️ Mikaël nous parle de son projet coup de cœur #OpenSource !

    ➡️ Le principe :
    - On décroche un vieux téléphone
    - On écoute le menu
    - On numérote
    - On laisse un message, on écoute une blague, etc.
    - On raccroche
    - Et tout est enregistré

    🛠️ Et techniquement ?
    - Un #RaspberryPi
    - Une carte son
    - Trois câbles

    Pour les curieux, c’est par ici :
    🔗 github.com/mmourcia/audio-gues

    #DIY #Makers #SideProject #Geek #Vibecoding #Vintage #AudioGuestbook

  21. El equipo de desarrollo del popular emulador de PS3, RPCS3, toma medidas, principalmente con los colaboradores que emplean el vibe coding, ante el aumento del envío de parches con código basura generado por IA.

    #gaming
    #RPCS3
    #vibecoding
    #PS3

    notasrem.com/equipo-rpcs3-toma

  22. 「 The team behind RPCS3, the open-source PlayStation 3 emulator, took to X today to very kindly and civilly request that users “stop submitting AI slop code pull requests” to its GitHub page. Then they immediately proceeded to tell the AI-brain-rotted tech bros attempting to justify their vibe-coding nonsense to kick rocks in the replies, which is somewhat less civil but far more entertaining to read 」

    kotaku.com/playstation-3-emula

    #ps3 #aislop #vibecoding #opensource #retrogaming

  23. BMad hat 46.000 GitHub-Stars und will Vibe Coding durch Specs, KI-Agenten und Quality Gates ersetzen. 5 Agenten, 4 Phasen, und ein Workflow, der sich anfühlt wie Wasserfall mit Autopilot. Klingt gut, aber: zwei Monate Lernkurve, und bei kleinen Projekten das 15-Fache der Zeit.

    Wann sich das lohnt, wann nicht, und welche Alternativen es gibt:

    kiberblick.de/artikel/workflow

    #KI #BMad #VibeCoding #ClaudeCode #Cursor #AITools #Entwicklung

  24. Vibe Coding = développement par intuition. Avantages : rapidité, créativité, adaptabilité. Risques : manque reproductibilité, bugs complexes. Quand l'utiliser : prototypage, optimisation. #Développement #Tech #VibeCoding #Innovation #BonnesPratiques ... linkedin.com/posts/gabriel-cha

  25. Vibe Coding = développement par intuition. Avantages : rapidité, créativité, adaptabilité. Risques : manque reproductibilité, bugs complexes. Quand l'utiliser : prototypage, optimisation. #Développement #Tech #VibeCoding #Innovation #BonnesPratiques ... linkedin.com/posts/gabriel-cha

  26. Vibe Coding = développement par intuition. Avantages : rapidité, créativité, adaptabilité. Risques : manque reproductibilité, bugs complexes. Quand l'utiliser : prototypage, optimisation. #Développement #Tech #VibeCoding #Innovation #BonnesPratiques ... linkedin.com/posts/gabriel-cha

  27. Vibe Coding = développement par intuition. Avantages : rapidité, créativité, adaptabilité. Risques : manque reproductibilité, bugs complexes. Quand l'utiliser : prototypage, optimisation. #Développement #Tech #VibeCoding #Innovation #BonnesPratiques ... linkedin.com/posts/gabriel-cha

  28. Vibe Coding = développement par intuition. Avantages : rapidité, créativité, adaptabilité. Risques : manque reproductibilité, bugs complexes. Quand l'utiliser : prototypage, optimisation. #Développement #Tech #VibeCoding #Innovation #BonnesPratiques ... linkedin.com/posts/gabriel-cha

  29. Hochauflösende Schalenmodelle für den Chemieunterricht individuell erstellen. Das Tool bietet zahlreiche Optionen für Arbeitsblätter, Lernplattformen oder Leistungsüberprüfungen.

    bohr.margomius.com

    #ChemieEdu #MINT #EdTech #Lehrer #Chemie #Unterrichtsmaterial #OpenEducationalResources #OERde #KI #vibecoding

  30. 516 tools, one tab, all free, nothing uploaded.

    Tooltimate replaces the bookmark folder of sketchy single-purpose tool sites you've been collecting since 2017.

    arcade.pirillo.com/tooltimate.

    #WebApps #IndieDev #VibeCoding #DevTools #Productivity #AI