If you ever wondered how well does it actually work.
tl;dr: good news, you are not loosing your job for now
To be fair, the last message in those PRs state:
The agent was blocked by configuration issues from accessing the necessary dependencies to successfully build and test. Those are being fixed and we’ll continue experimenting.
So let’s wait and see 😄
More links on Reddit
That “please, fix it” followed by 95 comments of failing to fix it perfectly describes my real world attempts to use copilot agents or “vibe code”
It’s good at the autocomplete stuff, agents are not nearly there and I just don’t see how they will be with the current LLM design.
Most of those comments are actually just random people arguing about the merits of the experiment, not continued discussion with the bot.
Also, the bot is supposed to be able to run builds to verify its work, but is currently prevented from doing so by a firewall rule they’re trying to fix, so its feedback is limited to what the comments provide. Humans wouldn’t do great in that scenario either. (Not to say the AI is doing “great” here, just that we’re not actually seeing the best-case scenario yet.)
Fair, but it’s still similar to my experiences when I’ve given it access to make and my full code base.
As I said, I don’t think it’s there yet.