Malicious compliance should be the order of the day. Just approve the requests without reviewing them and wait until management blinks when Microsoft's entire tech stack is on fire. Then quit your job and become a troubleshooter on x3 the pay.
Watching AI drive Microsoft employees insane
21–30 of 580 posts
Re: Watching AI drive Microsoft employees insane
#22Do we know for a fact there are Microsoft employees who were told they have to use CoPilot and review its change suggestions on projects? We have the option to use GitHub CoPilot on code reviews and it’s comically bad and unhelpful. There isn’t a single member of my team who find it useful for anything other than identifying typos.
Depends on team but seems management is pushing it from https://news.ycombinator.com/item?id=44031432 "From talking to colleagues at Microsoft it's a very management-driven push, not developer-driven. Friend on an Azure team had a team member who was nearly put on a PIP because they refused to install the internal AI coding assistant. Every manager has "number of developers using AI" as an OKR, but anecdotally most d…
> Depends on team but seems management is pushing it
The graphic "Internal structure of tech companies" comes to mind, given if true, would explain why the process/workflow is so different between the teams at Microsoft: https://i.imgur.com/WQiuIIB.png
Imagine the Copilot team has a KPI about usage, matching the company OKRs or whatever about making sure the world is using Microsoft's AI enough, so they have a mandate/leverage to get the other teams to use it regardless of if it's helping or not.
Re: Watching AI drive Microsoft employees insane
#23I do love one bot asking another bot to sign a CLA! - https://github.com/dotnet/runtime/pull/115732#issuecomment-2...
Re: Watching AI drive Microsoft employees insane
#24So, to achieve parity, they should allow humans to also commit code without checking that it at least compiles, right?
Or MS already does that?
Re: Watching AI drive Microsoft employees insane
#25With how stochastic the process is it makes it basically unusable for any large scale task. What's the plan? To roll the dice until the answer pops up? That would be maybe viable if there was a way to automatically evaluate it 100% but with a human in the loop required it becomes untenable.
The plan is to improve AI agents from their current ~intern level to a level of a good engineer.
Seems like that is taking a very long time, on top of some very grandiose promises being delivered today.
Re: Watching AI drive Microsoft employees insane
#26Earlier quoted context omitted.
This is the future envisioned by Microsoft. Vibe coding all the way down, social network style. They are putting this in front of the developers as take it or leave it deal. I left the platform, doing my coding old way, hosting it somewhere else. Discoverability? I don't care. I'm coding it for myself and hosting in the open. If somebody finds it, nice. Otherwise, mneh.
As long as the resulting PR is less than 100 lines and the AI is a bit more self sufficient (like actually making sure tests pass before "pushing") it would be ok I think. I think this process is intended for fixing papercuts rather than building anything involved. It just isn't good enough yet.
Yeah, just treat it like a slightly more capable dependabot.
Re: Watching AI drive Microsoft employees insane
#27> This seems like it's fixing the symptom rather than the underlying issue?
Exactly. LLM does not know how to use a debugger. LLM does not have runtime contexts.
For all we know, the LLM could’ve fixed the issue simply by commenting out the assertions or sanity checks and everything seemed fine and dandy until every client’s device catches on fire.
Re: Watching AI drive Microsoft employees insane
#28Interesting that every comment has "Help improve Copilot by leaving feedback using the or buttons" suffix, yet none of the comments received any feedback, either positive or negative. > This seems like it's fixing the symptom rather than the underlying issue? This is also my experience when you haven't setup a proper system prompt to address this for everything an LLM does. Funniest PRs are the ones that "resolves" t…
> How are people reviewing that? 90% of the page height is taken up by "Check failure", Typically, you wouldn't bother manually reviewing something until the automated checks have passed.
I dunno, when I review code, I don't review what's automatically checked anyways, but thinking about the change/diff in a broader context, and whatever isn't automatically checked. And the earlier you can steer people in the right direction, the better. But maybe this isn't the typical workflow.
Re: Watching AI drive Microsoft employees insane
#29Do we know for a fact there are Microsoft employees who were told they have to use CoPilot and review its change suggestions on projects? We have the option to use GitHub CoPilot on code reviews and it’s comically bad and unhelpful. There isn’t a single member of my team who find it useful for anything other than identifying typos.
Depends on team but seems management is pushing it from https://news.ycombinator.com/item?id=44031432 "From talking to colleagues at Microsoft it's a very management-driven push, not developer-driven. Friend on an Azure team had a team member who was nearly put on a PIP because they refused to install the internal AI coding assistant. Every manager has "number of developers using AI" as an OKR, but anecdotally most d…
It kinda makes sense for management to push it. Nothing else has a hope of preventing MSFT's stock price from collapsing into bluechip territory.
Re: Watching AI drive Microsoft employees insane
#30Malicious compliance should be the order of the day. Just approve the requests without reviewing them and wait until management blinks when Microsoft's entire tech stack is on fire. Then quit your job and become a troubleshooter on x3 the pay.
> when Microsoft's entire tech stack is on fire
Too late?