My mother and her husband are COBOL devs for a US state government. She works on the health insurance side for teachers and other state employees. Think claim processing.
Lots of batch jobs running at night. Their alert system is an actual human who calls my mom when jobs fail in the middle of the night.
It's high paying for the city they live in, but not high paying for software development. They will both have full retirement and healthcare for life, assuming the government can fulfill it. They are both fully remote since COVID too.
She's also worked for state lottery, teacher's retirement system and DOT.
edit: she says they have a SQL database, but mostly store in IBM IMS
She says it's both. By new stuff, it's mostly one off programs that handle small changes to the way billing or claims are handled. It sounds like they have a library of programs to start with and she extends it to fit the new edge case.
Just re-wrote an internal CLI to urfave from custom, but having issues with v3 autocomplete scripts. Might just take the time to switch over to cobra and use this.
from my experience of late, engineers usually toss stack traces into LLMs and ask what the issue is. good error wrapping, engineers grep for one of the first errors and get to the error code right away. both ways work.
Counterpoint: You might also just be working on an exceptionally bad codebase or with an exceptionally bad runtime (I'm looking at you C/C++ or minified Javascript)
Sure, if the stack trace doesn't have any useful information then it's not useful.
In my experience it's usually useful though, it usually tells you both what the problem is and where it is. Sadly a lot of people just don't even try to read them.
I would guess most people in leadership have no actual clue how they would replace the human in the seat with AI. They are riding the hype that the companies who provide the tools are putting out, but when it comes time to execute they don't have a plan. The steps between generated code to running in production is completely being ignored.
Ironically, LLM’s seem most suited towards replacing many of the functions middle managers perform. But those folks would never allow their own role to be outsourced or automated away.
That's not really the dynamic. If director A used to oversee 3 teams, and she discovers that with AI she can now oversee 6 or 9 teams just as efficiently, she's not going to avoid doing it out solidarity with manager B who's looking for space to move up the ladder.
I'd say it's even more complex. Middle-level management exists because CXOs don't want to deal with their employees but at the same time want them to under control. I don't believe CXOs can trust machines to be as efficient at controlling people as managers. And then CXOs would have to deal with these machines that control people which introduces a new class of problems.
> have no actual clue how they would replace the human in the seat with AI
That's not how it works. It's not "your job is done by an AI now", it's "you have an AI to help with these specific tasks, so the team is now half the size and you each get twice as much work".
Oh interested in the issues with http clients you have? I think they are great to use for being right in the standard library. Only complaint I have is defaults for timeouts and such, but you make that mistake once and don't forget.
Related to the mastering your IDE, stop reaching for your mouse for every action. The devs that I view as the slowest in my org, it's always watching them drag the mouse around clicking for their next action. Scrolling a file tree looking for a file instead of fuzzy searching the name.
With the consequence that Americans, for instance, would be unlikely to vote themselves into paying $2.8 trillion annually for this more leisurely life. Or even to vote that other, richer Americans should pay it, because if less wealth is being extracted then the sustainability of the scheme looks a little sketchy.
On the other hand voters don't think much about long-term sustainability, but do think a lot about being given money, so this scheme would be politically viable regardless.
Lots of batch jobs running at night. Their alert system is an actual human who calls my mom when jobs fail in the middle of the night.
It's high paying for the city they live in, but not high paying for software development. They will both have full retirement and healthcare for life, assuming the government can fulfill it. They are both fully remote since COVID too.
She's also worked for state lottery, teacher's retirement system and DOT.
edit: she says they have a SQL database, but mostly store in IBM IMS