Yep! We have a review process where we have a few agents, each tuned to a particular domain of expertise (security, code quality, etc) which iterate until the feedback meets a certain threshold, at which point it goes over to humans for (hopefully) final review.
That said, I generally agree that you're correct: writing code in many ways has not been the biggest bottleneck. However, by removing much of that writing, it frees up engineers to work on the uniquely human things that are larger bottlenecks.
I had a few comments in a thread here touching on where I think most of the value has come from for us (which is largely search/understanding of our dependencies and making away team work far more viable, which aids with cutting through bureaucracy and the tendency for teams to push back on work): https://news.ycombinator.com/item?id=48298731
Haven't you heard - these days they just throw slop generated by LLM agents over to other LLM agents which cosplay as internal QA. They know it works because they write really strict .MD files where they instruct agents in English language to 'never do this' and 'always do that'.
This is really what you think happens at large tech companies? You don't think it's possible this is maybe even slightly overly simplifying what the relevant processes are?
Comment does indicate you don’t really seek to know how things work with respect to this and seem to not be able to imagine that the Occam’s razor is: agents are more useful than you think they are.
I would not agree on that point. It really depends on company's structure. I mean it also depends with people that makes the team. I would say there are a lot of unknowns but I would certainly not generalize.
How I find your argument is that one distinguished engineer from US could do the same with the use of AI.
I worked with both and I know great and bad engineers from both sides. Only thing is that US has a bigger pool of great engineers.
I am not sure this feels right. I agree that the US currently has leading minds in terms of tech, but I am not sure it is too big of difference with the EU knowledge workers. EU is still a lot cheaper then US in terms of wages you would need to pay.
EU workers themselves get a lot less, but the EU is expensive because of 1) the huge payroll tax (45% in France) and 2) the challenges with hiring and firing mean you are carrying people that aren’t contributing.
Sure is an interesting thought. None of this is sarcasm: why do US companies deal with the time zone differences and language barriers they won’t need to bother with so much by outsourcing to say, Ireland?
The mechanism is often that they'll actually outsource to someone like Accenture, who have teams everywhere, and whose contract managers will try to get their cheapest viable team onto the contract to maximise their margin. If the buyer can't judge the quality of what they're buying, or doesn't know why the resulting hand-offs, delays, mistakes and rework will cost them more than keeping everything in-house ever would have, they're going to have a bad time.
Basically every big tech has large offices and employ a lot of people there.
The limitation is that Ireland is a relatively small country, and most Irish developers are already employed (which is why Ireland end up being one of the main destinations for tech workers being hired from abroad).
Ireland isn't that much cheaper than, say, Oklahoma. And the cultural differences with Ireland are not a lot smaller than those with India or the Philippines or what have you, once you try to actually start working together.
(Yes, all the good developers from Oklahoma move out, but the same is true of Ireland)
I agree that most of the things are written by AI but writting code was never the bottleneck in big tech.
reply