In my org, it's grown the level of work. We had a lot of stuff that was never worth the devtime necessary, but now that's opened up that we can do a lot of this stuff in the background
I think this is underrepresented in everyone's calculations about how AI will affect software engineering. In my experience (and apparently yours as well) this is what many companies are using it for. All those pesky bugs that are minor annoyances but not show-stoppers are getting addressed. They aren't helping us sell more software, but they're dissatisfiers for the customers.
Worth looking into 'Jevons Paradox' for more examples of this sort of thing. TL;DR is that as things become more efficient, whether coal, or programmers, the usage of a good increases as demand skyrockets for what it can be used for.
But it's at my day job, and it's because I was able to write a prompt which automates having Copilot review uploaded scanned PDFs of invoices with checks (and the bank line obscured with a pen, so no PII) and then write a batch file which renames the files per a file-naming convention, removing the need to open them in batches of 50, find the Invoice ID, re-save using that filename, then quit and re-launch Adobe Acrobat (if left running, eventually I run into a bug where it stops saving files), then run a .bat file which renames based on Invoice ID as a filename.
Problem of course is I've been running into a limit of number of allowed files per 24 hr. period.
Even if it's not less work, it feels like less effort.
Extracting structured data from PDFs is something people have been doing forever, way before LLMs, there are state of the art OCR tools out there you can use as a lib, it’s hardly something that justify using a LLM
The position of (for instance) the check # on a physical check varies quite widely, as can date position and format, and a fair number of them are still written by hand.
On top of that, I'm scanning thousands of these each year, and the invoice underneath the check ranges from the gamut of: "pristine copy just re-printed 'cause the customer didn't include one" through "bad inkjet photocopy of a photograph taken w/ a potato phone and then printed" and includes variations such as "customer included half-a-dozen invoices to be paid w/ one check, and if arranging them and the check _just so_ all will fit in the document camera window, saving a trip to the sheet feeder scanner"
A co-worker actually worked on this for a different program, one where actually sending in paperwork in good condition was expected and customary, but his system had a reject/failure rate of ~10% --- that would be almost 1,000 invoices each year for the program I am handling.
Yes it is. It's a poorly defined problem, and it needs to be correct, and existing OCR tools aren't perfect. Are the end of the day, it comes down to money and time, and throwing them at ChatGPT is cheap and fast enough to be used.
I have the opposite, because I'm now getting things clients generated that they want me to implement. It's definitely more work and money for me, but it's concerning somewhat because ultimately it's not good for their business. It takes me longer to implement this kind of stuff than it would for me to code it from scratch. That is in a working way, the AI generated code has far more bells and whistles, but also layers and layers of needless complexity that quite literally add no value as in they aren't even a factor of the finished output.
The problem is they are now paying me more, plus paying for the cost of using the AI, and the needless complexity also slows down the employees. So more costs there as well, any future debugging is going to cost far more and at the end of the day they are getting less quality on the core function but far more presentation data that is essentially meaningless.
Nope, but recently it does feel like I spend less time coding and more on prep work / requirements gathering. But that's n=1, I'm doing something very specific (building a design system with a small team of designers and developers) and recently my skill in writing requirements and doing research has been recognised.
But I'm trying to figure out if this is a thing in my organization. So far the channels discussing AI have been about how they use AI, how to build agents, and complaining they've run out of tokens - but nobody says anything about how much work they finished, how it impacted the product(s), the user, and the company's bottom line.
I don't know if they just don't know, don't care, don't measure, or they don't actually want to admit they're just burning through tokens with nothing to show for it.
That said, right now I'm using AI to extract some code from our app to create a minimal reproduction for an issue we're having with a 3rd party library, it's a huge time saver there - that is, I wouldn't have bothered making a reproduction because of how time consuming it can be.
So to answer your question, it's not that "I have less to do", it's "I do things I wouldn't have done otherwise because they cost less effort/time". Which is also the promise of e.g. automation - the automated loom didn't just reduce how long it takes to make cloth, it made it so people make a bajillion times more cloth.
Yes, I have seen it happen on enterprise consulting.
For example, all the work like translations that used to be done by humans, now it is a CMS AI feature.
Secondly teams setup.
It used to be we did everything ourselves for development, then cloud, SaaS products and serverless decreased the teams size required for delivery.
Now with AI, there is an even greater push for low code/no code tooling, with agents, leaving the actual programming left for MCP tools that might not yet be available for the project.
Thus you get a team of five doing what used to be about 15 a decade ago.
I dunno I type a whole bunch of shit into a computer and software comes out the other end. The experience of doing the work hasn’t changed much but I get better software more quickly.
The main thing that matters is who takes responsibility for it. And the other thing is honesty, e.g. "I made this myself" vs "I made this with the help of AI". I'm a bit precious about that though, like, I wouldn't really care if someone used vim or emacs to write code not long ago - the tools didn't matter.
If you are asking if the machine translated from one language to another for him, the answer is essentially guaranteed to be yes. Inputting raw machine code hasn't been the norm since the 1950s.
I wonder if there is something like translation party, only for compilers and LLM respectively. There should be!
I.e. some source code gets compiled, the output gets decompiled again, the result gets compiled again, and so on.
Now do the same with LLM: start with a prompt, it generates a program. Then an LLM examines that program by looking at the source, running it, etc., and is tasked to describe it, i.e. turn it into a prompt again. Then that prompt is used to generate a program, which is examined and turned into a prompt again, and so on.
Compilers and assemblers are not the same either, but the human experience is the same in all cases: You input one language into the machine and another language comes out. Implementation details are immaterial.
That is because you only look at one hypothetical experience, in a vacuum, not the continued experience. In other words, removing the dimension of time, and the fact that you get wildly different output for the same input with LLM, not so for compilers.
It's like a light switch where the light has a slightly different colour every time you turn it on, sometimes the colour is very different, and sometimes it emits bursts of heavy gamma radiation. And the argument is "you flick a switch, and what you perceive in that room changes, it's the same experience." (which is a neat way to put it, because it's true even when you get evaporated by the gamma radiation burst)
That this isn't "the same kind of tool" than just a light switch and a bulb isn't even an interesting conversation to have. Of course it's not, not even close. A piece of string is not a steel rod just because you can stretch it out, take a photo of it, and play pretend.
So the really interesting bit to me is that this level of argument is even made. It's nonsense, but if the nonsense gets repeated often enough, maybe the people refuting it can be exhausted and then we can replace concrete with pudding, right? Not that this is your intent, but this stuff just feels like the thing you'd hear in a digestive tract, not at a table between peers.