The study tracked around 800 developers, comparing their output with and without GitHub’s Copilot coding assistant over three-month periods. Surprisingly, when measuring key metrics like pull request cycle time and throughput, Uplevel found no meaningful improvements for those using Copilot.

  • Takumidesh@lemmy.world
    link
    fedilink
    English
    arrow-up
    22
    ·
    2 months ago

    I basically exclusively use LLMs to explain broad concepts I’m unfamiliar with. a contrived example would be ‘what is a component in angular’ or ‘explain to a c# dev how x,y, and z work in rust’ The answers don’t need to be 100% accurate and they provide a nice general jumping point to get specific information.

    • xavier_berthiaume
      link
      fedilink
      English
      arrow-up
      5
      ·
      2 months ago

      Exactly, I’ve found them most useful to either summarize a text you feed it, or do broad ‘google like’ queries. I don’t trust it with anything beyond that