I ask it to do something, if it works, I keep it without looking at it. If it doesn't I look, recoil in horror, delete and rewrite myself.
It takes more time but seems to work for me.
461–470 of 744 posts
I ask it to do something, if it works, I keep it without looking at it. If it doesn't I look, recoil in horror, delete and rewrite myself.
It takes more time but seems to work for me.
Earlier quoted context omitted.
Yes, but for experienced engineers that is still a huge huge change . Even 12 months ago simplifying tasks alone was insufficient, you still needed a large group engineers to actually write, review and maintain a typical product for solid startup offering. This came with the associated overhead of hiring and running mid sized teams. A lot of skilled people (y)our age/experience are forced into doing people management…
The funny thing is a lot of that was never really necessary per se. Tons of stories about great projects coming out of tiny teams. They’re not likely geniuses, they just had focus and clarity and a drive to GSD without excessive unproductive activity. I’ve long been a proponent of offshore developers for cost savings. You have to manage the process and people differently, but the output per dollar (pre AI) was phenom…
I have noticed lately that getting into an USA company as a foreigner became very difficult. I get a lot of praise on culture fit and tech assignments and then get told off with something very similar to "can't get compliance to agree to work with Bulgaria".
Sigh.
But I get what you mean. I started using LLMs and that gave me a perspective what it is to be an engineering manager.
> I'm sure every organisation has hundreds if not thousands of Excel sheets tracking important business processes that would be far better off as a SaaS app. Far better off for who? People constantly dismiss spreadsheets, but in many cases, they are more powerful, more easily used by the people who have the domain knowledge required to properly implement calculations or workflow, and are more or less universally acce…
It’s a rare spreadsheet that survives its original creator.
The cost of writing simple code has dropped 90%. If you can reduce a problem to a point where it can be solved by simple code you can get the rest of the solution very quickly. Reducing a problem to a point where it can be solved with simple code takes a lot of skill and experience and is generally still quite a time-consuming process.
Well said. The cost of building a CRUD has dropped 90%. The open question is why people needed fancy AI tools like Claude to write CRUDs in the first place. These kind of tasks ought to be have been automated a long time ago.
The number of lines of code to be written is much, much lower than in the early 2000's.
just thinking from the finance world, in 2010 no knew how to program on the desk and no one knew sql, and even if they did the institutional knowledge to use the dev systems was not worth their time. So you had multiple layers of meetings and managers to communicate programs. As a result, anything small just didn't get done, and everything took time.
by 2020 most junior guys knew enough python to pick up some of the small stuff
in 2025 ai tools are good enough that they're picking up things that legit would have taken weeks to do in 2010 because of the processes around them not the difficulty and doing it in hours. A task that would take an hour to do used to take multiple meetings to properly outline to someone without finance knowledge and now they can do themselves in less time than it took to describe to a fresh cs grad.
Those tasks that junior traders/strats are able to do now that would have taken weeks or months to get into prod going through an it department i'm seeing cost drop 90% everyday right now. Which is good, it lets tech focus on tech and not learning the minutia of options trading in some random country
The cost of writing simple code has dropped 90%. If you can reduce a problem to a point where it can be solved by simple code you can get the rest of the solution very quickly. Reducing a problem to a point where it can be solved with simple code takes a lot of skill and experience and is generally still quite a time-consuming process.
Well said. The cost of building a CRUD has dropped 90%. The open question is why people needed fancy AI tools like Claude to write CRUDs in the first place. These kind of tasks ought to be have been automated a long time ago.
They have been, repeatedly, since the 70s. See dBase, Clipper, Microsoft Access, Hypercard, Ruby on Rails, stretching Wordpress to within an inch of its life, all manner of "no-code" things...
And, honestly, Excel. People do all manner of terrifying things with Excel, and it is unquestionably the most successful, and arguably the _only_ successful, "we can do this thing instead of employing a programmer" tool.
Generally, one of two things has happened. Either (a) the products of such automation become unmaintainable nightmares (common for the more automated approaches like MS Access) or (b) they become complex enough that they tend towards 'normal' programming (common with, say, Rails, where you could get a simple CRUD with basically just DSL, but realistically eventually you're gonna be writing lots of Ruby).
I feel like LLM-produced stuff is probably going to fall into column A.
Earlier quoted context omitted.
> The cost of writing simple code has dropped 90%. Plus there's a lot of simple code you shouldn't be writing either way , because it's in a library by now. By their nature, LLMs will do their best with things that could be plagiarized.
Aren't we having major issues with there being too many small libraries right now and dependency chain that grows exponentially? I have thought LLMs will actually benefit us a lot here, with not having to use a lib for every little thing (leftpad etc?).
(IMO Javascript desperately needs an equivalent to Boost, or at the very least something like Apache Commons.)
Earlier quoted context omitted.
For me, as a remote developer, it means I'm able to finish my work in 1 hour instead of 8 hours. So I'm able to capture "extra value" in the form of time. In our team everyone uses GitHub Copilot and I use Claude Code. My teammates' productivity increased slightly but my productivity increased a lot. This is because 1. Claude Code is just a better coding agent 2. I invested time to get good at agentic coding. Eventua…
I'm really curious on what your role is, and which industry are you in? I'm awed by these productivity gains others report, but I feel like AI helps in such a small part of my job (implementing specific changes as I direct). Agentic workflows for me results in bloated code, which is fine when I'm willing to hand over an subsystem to the agent, such as a frontend on a side project and have it vibe code the entire thin…
Had the cost of building custom software dropped 90%, we would be seeing a flurry of low-cost, decent-quality SaaS offering all over the marketplace, possibly undercutting some established players. From where I sit, right now, this does not seem to be the case. This is as if writing down the code is not the biggest problem, or the biggest time sink, of building software.
Well, because no self interested decision maker in any company of size is going to ever trust their business to an unknown company run by a one person operation. And why would the benefits of being able to code faster accrue to a small independent developer over a large company that already has an established reputation and a customer base? “No one ever got fired for buying Salesforce”. I once had influence over the…