home.social

#codereviews — Public Fediverse posts

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

fetched live
  1. I've built a thing:
    zed.dev/extensions/word-highli

    It's a @zed #lsp #extension to highlight all occurrences of selected words with multiple colors to quickly identify them in code.

    I find it useful for #codereviews. Enjoy!

  2. I've built a thing:
    zed.dev/extensions/word-highli

    It's a @zed #lsp #extension to highlight all occurrences of selected words with multiple colors to quickly identify them in code.

    I find it useful for #codereviews. Enjoy!

  3. "First, the easy and obvious one: code review. CTOs keep asking me if code review is really dead. The answer: Not Yet. But it will be by next year. You can't work at agentic speeds and block everything with human reviews. Those are incompatible. Your competitors are going to be moving at agentic speeds, so you're fucked in the medium-term if you stick with human code reviews. That's about as bare as I can make it for you.

    Whenever someone says code reviews are going away, someone else says "SOC 2." Somehow SOC 2 spread through the industry without any actual laws being passed, as far as I can tell, which is bizarre. But big enterprise clients will require their vendors to be SOC 2 compliant, for auditing purposes. And many people interpret it as requiring human code reviews.

    As a result, human approval is currently baked into many companies' audited change-management controls and customer commitments. But the writing is on the wall: agentic throughput will straight-up force those controls to be rewritten. SOC 2 will no doubt survive, but "review" will no longer mean one human approving every diff.

    Upshot: in the short term, yes, keep reviewing agent code. Fable is the only reasonably trustworthy model in existence today, and you're not going to want to use it much due to its exorbitant pricing. But in seven months, all the models will be that smart, and inference will be much cheaper. Plan now for that day to arrive, and get ready to replace your human code review with many, many rounds of agentic code review.

    It will go better for you anyway. Humans suuuuck at code review. I've been doing this for forty years, and all I've seen is decades of thinly-disguised LGTMs. Just let it go, mate. It's almost time."

    yegge.ai/essays/the-shape-of-t

    #AI #LLMs #GenerativeAI #SoftwareDevelopment #Programming #CodeReviews #CICD

  4. "First, the easy and obvious one: code review. CTOs keep asking me if code review is really dead. The answer: Not Yet. But it will be by next year. You can't work at agentic speeds and block everything with human reviews. Those are incompatible. Your competitors are going to be moving at agentic speeds, so you're fucked in the medium-term if you stick with human code reviews. That's about as bare as I can make it for you.

    Whenever someone says code reviews are going away, someone else says "SOC 2." Somehow SOC 2 spread through the industry without any actual laws being passed, as far as I can tell, which is bizarre. But big enterprise clients will require their vendors to be SOC 2 compliant, for auditing purposes. And many people interpret it as requiring human code reviews.

    As a result, human approval is currently baked into many companies' audited change-management controls and customer commitments. But the writing is on the wall: agentic throughput will straight-up force those controls to be rewritten. SOC 2 will no doubt survive, but "review" will no longer mean one human approving every diff.

    Upshot: in the short term, yes, keep reviewing agent code. Fable is the only reasonably trustworthy model in existence today, and you're not going to want to use it much due to its exorbitant pricing. But in seven months, all the models will be that smart, and inference will be much cheaper. Plan now for that day to arrive, and get ready to replace your human code review with many, many rounds of agentic code review.

    It will go better for you anyway. Humans suuuuck at code review. I've been doing this for forty years, and all I've seen is decades of thinly-disguised LGTMs. Just let it go, mate. It's almost time."

    yegge.ai/essays/the-shape-of-t

    #AI #LLMs #GenerativeAI #SoftwareDevelopment #Programming #CodeReviews #CICD

  5. This is a source of concern with us as we transition to AI-assisted/AI-first development. The linked article could do with some editing, but it has some interesting points. But one thing is clear, traditional PR code-reviews are just not fit for purpose anymore!

    #AI #LLM #CodeReviews #SDLC #Coding #SoftwareDevelopment #ThoughtWorks

    thoughtworks.com/insights/blog

  6. This is a source of concern with us as we transition to AI-assisted/AI-first development. The linked article could do with some editing, but it has some interesting points. But one thing is clear, traditional PR code-reviews are just not fit for purpose anymore!

    #AI #LLM #CodeReviews #SDLC #Coding #SoftwareDevelopment #ThoughtWorks

    thoughtworks.com/insights/blog

  7. RE: social.wildeboer.net/@jwildebo

    Wenn jemand noch Gedanken und Erfahrungen zu Code Reviews / Pull Requests von mit LLMs erzeugtem Code hat, bitte hier entlang! @jwildeboer freut sich bestimmt über weiteren Input für seinen weiter gehenden Blogpost.

    #AI #KI #CodeReviews #Softwareentwicklung

  8. RE: social.wildeboer.net/@jwildebo

    Wenn jemand noch Gedanken und Erfahrungen zu Code Reviews / Pull Requests von mit LLMs erzeugtem Code hat, bitte hier entlang! @jwildeboer freut sich bestimmt über weiteren Input für seinen weiter gehenden Blogpost.

    #AI #KI #CodeReviews #Softwareentwicklung

  9. Mark Dominus thinks folks are clueless about code reviews, but let's face it, it's just another excuse for programmers to inflate their egos and nitpick syntax like they're the grammar police of the digital world. 🤓👉🧐 Meanwhile, #Mastodon users are told to embrace #JavaScript or go find an app, proving once again that the simplicity of the Stone Age is long gone. 🦕💾
    mathstodon.xyz/@mjd/1150967203 #codeReviews #egoInflation #digitalSimplicity #HackerNews #ngated

  10. Mark Dominus thinks folks are clueless about code reviews, but let's face it, it's just another excuse for programmers to inflate their egos and nitpick syntax like they're the grammar police of the digital world. 🤓👉🧐 Meanwhile, #Mastodon users are told to embrace #JavaScript or go find an app, proving once again that the simplicity of the Stone Age is long gone. 🦕💾
    mathstodon.xyz/@mjd/1150967203 #codeReviews #egoInflation #digitalSimplicity #HackerNews #ngated

  11. Meet Pullfrog - an #opensource AI-powered GitHub bot designed for automation in GitHub Actions.

    Key features include orchestration for pull request reviews, issue triage, and CI remediation, all managed within GitHub’s environment.

    The tool operates with a bring-your-own-key model for access.

    Details here 👉 bit.ly/4x1vLN2

    #InfoQ #WebDevelopment #CodeReviews #AIAssistedCoding #AIDevelopment

  12. Meet Pullfrog - an AI-powered GitHub bot designed for automation in GitHub Actions.

    Key features include orchestration for pull request reviews, issue triage, and CI remediation, all managed within GitHub’s environment.

    The tool operates with a bring-your-own-key model for access.

    Details here 👉 bit.ly/4x1vLN2

  13. Discover how #Meta improved #SoftwareQuality with a Just-in-Time (JiT) testing approach that dynamically generates tests during code review.

    The system increases bug detection by approx 4x in AI-assisted development using LLMs, mutation testing, and intent-aware workflows like Dodgy Diff.

    More on #InfoQbit.ly/4tqylKc

    #SoftwareArchitecture #SoftwareTesting #AI #LLMs #CodeReviews

  14. Discover how improved with a Just-in-Time (JiT) testing approach that dynamically generates tests during code review.

    The system increases bug detection by approx 4x in AI-assisted development using LLMs, mutation testing, and intent-aware workflows like Dodgy Diff.

    More on bit.ly/4tqylKc