But you'll use the mediocre software written by hundreds of mediocre devs, and a team that doesn't fully understand the whole codebase due it's ever-inflated technical debt?
I get the angst of AI coding, but it is funny that people think hand written code at scale is somehow not susceptible to all of the problems of code gen. A lot of it isn't artful - it's factory generated slop.
Isn't this somewhat semantical? Boring implies a lot of the things you said, especially if it means your team's experience is largely pooled in a particular dev environment. Assuming most boring tech is ubiquitous, it's probably rare that your team, statistically, is deeply literate in some obscure tech - they most likely are experts in some definition of "boring".
TLDR; #1 and #2 are essentially implying the same thing.
insanitybit's point is that "boring" can imply those things, but it can imply other things too. If you have to bring the correct context to make the word make sense, and there's other contexts that could apply that don't make sense, the word isn't helping. You bringing the correct context is doing the heavy lifting.
I guess context clues and the way the post was written had obvious implications - that was my point. I do understand it leaves a little open to interpretation, which is the angle he decided to comment on.
Is it long term pain if a small startup in a very competitive vertical gets to market quicker with code gen and finds market fit? Whether that results in a rewrite or a refactor, that is still a huge win and one that a lot of people will gamble on.
Example of startups/incumbents that are using AI to reinforce workflows? I am assuming you don't read up on this stuff much, but it's everywhere. I can give you multiple examples in my industry (logistics) that are actively using AI to speed up doc collection, invoicing, maintenance cycles, and in general, a lot of back office flows.
These are just some of the bigger guys that are either raising seed funding or already well-established... I can't imagine all the small and obscure teams that are chasing the same thing.
A lot of these guys have marketshare already and are weaving AI surgically through the stack, which is how people should be building. These aren't AI wrapper companies... these are companies that are leveraging AI to amplify their existing products.
I am not sure why it's so hard for people to assume the semantical analysis that AI offers can't be leveraged properly inside of a product.
Take this one post, and then extrapolate similar products across all verticals... that is currently the waters that are being tested. Law, healthcare, construction, etc - there all getting pressure tested for similar AI integrations.
But if the code base is constructed in a way which results in strong test suites (human driven), quality evals, robust documentation, and modular implementation, will the human really lose that much context?
Given that those things are checked off, I am having a tough time believing a frontier model, in conjunction with a new dev on a project, can't be immediately valuable to determining what is going on in a repo.
Maybe this is an elementary angle given my lack of security experience, but couldn't Microsoft figure out a wat to parse the documents prior to model analysis/action? Implement some form of deterministic layer that resides between the user and the model?
Parse for what? The model has “arbitrary understanding” of “arbitrary input”. The filter is unbounded and the only actually safe result is to filter everything.
I get the angst of AI coding, but it is funny that people think hand written code at scale is somehow not susceptible to all of the problems of code gen. A lot of it isn't artful - it's factory generated slop.
reply