[–] 14 points 10 months ago (5 children)

A task that might have taken five hours assisted by AI, and perhaps ten hours without it, is now more commonly taking seven or eight hours, or even longer.

What kind of work do they do?

in my role as CEO of Carrington Labs, a provider of predictive-analytics risk models for lenders. My team has a sandbox where we create, deploy, and run AI-generated code without a human in the loop. We use them to extract useful features for model construction, a natural-selection approach to feature development.

I wonder what I have to imagine this is doing and how. How do they interface with the loop-without-a-human?

Either way, they do seem to have a (small, narrow) systematic test case and the product variance to be useful at least anecdotally/for a sample case.

  • source
  • [–] 1 point 10 months ago

    I am happy to share that we (the @GoogleAIStudio team) are now a sponsor of the @tailwindcss project! Honored to support and find ways to do more together to help the ecosystem of builders.

    by Logan Kilpatrick - user profile desc: Lead product for @GoogleAIStudio & Gemini API. My views!

  • source
  • [–] 5 points 10 months ago

    In December 2024, the BBC carried out research into the accuracy of four prominent AI assistants that can search the internet – OpenAI’s ChatGPT; Microsoft’s Copilot; Google’s Gemini; and Perplexity. We did this by reviewing responses from the AI assistants to 100 questions about the news, asking AI assistants to use BBC News sources where possible.

    The answers produced by the AI assistants contained significant inaccuracies and distorted content from the BBC. In particular: …

    51 % significant issues, 19 % factual errors, 13 % altered or invalid quote citations

  • source
  • [–] 1 point 10 months ago

    The article isn't very concrete on what the substance of this Anyway System is.

    The FAQ answered my question though, and it seems mostly about dynamic management of PCs as execution nodes.

    I assume that means I could run the model on one of those PCs as well, despite the article claiming you can use as few as four PCs? Or does this system somehow distribute a single model prompt process into multiple execution pipelines?

  • source
  • [–] 2 points 10 months ago

    Without understanding how it’s built, how do I know if there’s duplication, dead code, or poor patterns? I used to obsess over this. Now I’m less worried that a human needs to read the code, because I’m genuinely not sure that they do.

    What you do need: simple entry points, explicit code with fewer abstractions, minimal coupling, and linear control flow.

    Seems to be the common simple standard software works well fallacy.

    By “can replace developers”, what do they mean? They don't clarify, only talk about their three success projects.

    We've seen studies of the issues and risks, and discrepancy between user perception and more factual gains. And this post certainly seems like they're not experienced in or thinking of development and maintenance that goes beyond simple standard integration software. Which doesn't make it too surprising they're not concerned about security for those simpler projects either.

  • source
  • [–] 1 point 10 months ago

    Modern websites no longer fit the document-centric model HTML was created for. A typical news homepage mixes headlines, images, teasers, and interactive elements in ways the original spec never anticipated. The New York Times even present teasers without headlines at all. This diversity shows how little shared foundation there is for developers today – and why HTML needs a broader, more coordinated evolution beyond isolated improvements.

    Aren't such cases already covered? I don't see the issue or alternative.

  • source
  • [–] 1 point 10 months ago (1 child)

    if the software developer had experience with AI

    Did these developers not have experience with AI?

    and were to start on a new project, without any existing context

    I'm not sure focusing on one aspect to scope a reasonable and doable study automatically makes it “really low effort”.

    If they were to test a range of project types, it'd have to be a much bigger study.

  • source
  • parent
  • context
  • [–] 9 points 10 months ago

    Before starting tasks, developers forecast that allowing AI will reduce completion time by 24%. After completing the study, developers estimate that allowing AI reduced completion time by 20%. Surprisingly, we find that allowing AI actually increases completion time by 19%--AI tooling slowed developers down.

    The gap and miss is insane.

  • source
  • parent
  • context
  • [–] [S] 29 points 10 months ago

    A relatively uncommon but reasonable, good approach to issue management.

    Discussions allow for different formats, including explicit voting, which is useful for things like feature requests.

  • source
  • view more: ‹ prev next ›