581–590 of 1001 posts
Earlier quoted context omitted.
Does it make you 25% more productive?
If 25% of code was AI-written, wouldn't it be a 33[.333...]% increase in productivity?
Wait, no. That should be based on how much slacking off Google employees do ordinarily, an unknown quantity.
I work for Google, and I just got done with my work day. I was just writing I guess what you'd call "AI generated code." But the code completion engine is basically just good at finishing the lines I'm writing. If I'm writing "function getAc..." it's smart enough to complete to "function getActionHandler()", and maybe suggest the correct arguments and a decent jsdoc comment. So basically, it's a helpful productivity…
In the hands of a junior, AI can create a false sense of confidence and it acts as a technical debt and security flaw multiplier.
We should bring back the title "Software engineer" instead of "Software developer." Many people from other engineering professions look down on software engineers as "Not real engineers" but that's because they have the same perspective on coding as typical management types have. They think all code is equal, it's unavoidable spaghetti. They think software design and architecture doesn't matter.
The problems a software engineer faces when building a software system are the same kinds of problems that a mechanical or electrical engineer faces when building any engine or system. It's about weighing up trade-offs and making a large number of nuanced technical decisions to ultimately meet operational requirements in the most efficient, cost-effective way possible.
Earlier quoted context omitted.
So your test is "seems to work"?
No, what I meant is that, much like when copying code, I only keep the generated source code if it's written the way I would write it. (By "unread" I meant that I don't look very closely before deleting if it looks weird.) And then write tests. Or perhaps I wrote the test first.
Earlier quoted context omitted.
Interesting that you believe your subjective experience outweighs the claims of all others who report successfully using LLMs for coding. Wouldn't a more charitable interpretation be that it doesn't fit the stuff you're doing?
Why wouldn't someone's subjective experience outweigh someone else's subjective experience? Regardless, I do wonder how accurate those successful reports are. Do people take LLM output, use it verbatim, not notice subtle bugs, and report that as success?
Earlier quoted context omitted.
I'm continually surprised by the amount of negativity that accompanies these sort of statements. The direction of travel is very clear - LLM based systems will be writing more and more code at all companies. I don't think this is a bad thing - if this can be accompanied by an increase in software quality, which is possible. Right now its very hit and miss and everyone has examples of LLMs producing buggy or ridiculou…
> The direction of travel is very clear And if we get 9 women we can produce a baby in a single month. There's no guarantee such progression will continue. Indeed, there's much more evidence it is coming to a a halt.
Earlier quoted context omitted.
> The direction of travel is very clear And if we get 9 women we can produce a baby in a single month. There's no guarantee such progression will continue. Indeed, there's much more evidence it is coming to a a halt.
Its not even been 2 years, and you think things are coming to a halt?
Earlier quoted context omitted.
I'm continually surprised by the amount of negativity that accompanies these sort of statements. The direction of travel is very clear - LLM based systems will be writing more and more code at all companies. I don't think this is a bad thing - if this can be accompanied by an increase in software quality, which is possible. Right now its very hit and miss and everyone has examples of LLMs producing buggy or ridiculou…
I'm not really seeing this direction of travel. I hear a lot of claims, but they are always 3rd person. I don't know or work with any engineers who rely heavily on these tools for productivity. I don't even see any convincing videos on Youtube. Just show me on engineer sitting down with theses tools for a couple hours and writing a feature that would normally take a couple of days. I'll believe it when I see it.
Why no youtube videos thought? Well, most dev you tubers are actual devs that cultivate an image of "I'm faster than LLM, I never re-read library references, I memorise them on first read" and do on. If they then show you a video how they forgot the syntax for this or that maven plugin config and how LLM fills it in 10s instead of a 5min Google search that makes them look less capable on their own. Why would they do that?