[–] [S] 1 point 2 months ago

The tool was made with AI (the community description says “people using AI for coding”), although I’m not using an agent. I’m fine with moderators (if there any) deleting if it’s not relevant.

  • source
  • parent
  • context
  • [–] [S] -1 points 2 months ago

    Note from OP: I posted this because I find a topic of "agent swarms" super interesting as seen as an engineering challenge. It's just a novel kind of problem yet to be solved. And this article describes what worked well and what not, what gains did they get making certain changes.

    But being honest I see some gaps in their reasoning. SQLite is an open source project which most probably means its code already is part of a training set of frontier LLMs. This calls into question whether language models actually invented or just recalled. Also, as noted in the article, they instructed models to build something based on the docs. That means that the goal was to recreate, not to invent. One of the hardest things in software engineering is to figure what to build and this agent swarm has already been given that information, what made the job much easier.

  • source
  • [–] 2 points 5 months ago (2 children)

    it seems to depend on how close you are; culturally speaking; to the anglosphere.

    look at rednote for an example of a large platform -- very distant from the anglosphere -- embracing ai content.

  • source
  • parent
  • context
  • [–] 2 points 5 months ago

    I have been working on setting up a reliable workflow using a combination of gemma4:26b, deep seek coder, and qwen3.5:27b

    I'm not a programmer is results are meh and I'm running into container issues but the speed is impressive and it's good enough that with some trouble shooting and verification between models I'm making progress.

  • source