It depends on the task. Oneshotting 80% of the new feature and then failing to pick up on details can leave hard part.
But oneshotting even 30% of UI tests based on QA regression means you have 30% of regression covered. Ofc you need to verify if the test is indeed covering the test case, and for a big volume of tests it takes time, but it's nowhere near what it takes to actually write the tests.
Pretty much my experience and I didn't even have to do workflows or complex harnesses, I wrote a "dont ask" mode wrapper that gives rights to read/write a work dir & explanation that blockers & questions need to be written in a specific directory, and I listen to this directory with a GUI app that notifies me, then also a stop hook that verifies that that doc is updated when it stops, and has all the items done/blocked.
I did this cause I like the flexibility of a normal agentic chat session.
Recent LLM are smart enough to resolve many problems as is in agentic mode. Really strange to see "it doesnt work" copium instead of fighting for the means of production and looking for a way to have this setup purely local.
It feels so weird to read lemmy, as if I'm living in a different reality. To me, most of the time an agent can oneshot a ticket (if it has a good, non-vague description) or at least do 80-90% that can be fixed with several changes or prompts, and only odd tasks need more manual investigations than that.
Surely you still need a dev oversight and someone needs to do tech plans & lead the projects, but that's not like anything people experience here.
I've not used gpt for quite a while, but can't you ask to python script the evaluation?
There is nothing unethical in the goal to simplify development. The only unethical thing here is how giant American closed source AI corpos try to trample the competition to become monopoly in providing essential tools.
AIs need to be open source and hardware to run AIs needs to be obtainable.
Your devs do QA, not a dedicated team?
Unit tests, snapshot tests, UI tests?
Why though? Aside from blind hate, good written JIRA stories are almost oneshot now in many occasions. You can spend more mental energy on logic correctness rather than translation of requirements into code.
Not only it is faster, but you can deliver greater quality faster. Slopware is a result of trying to crank the dev speed to the max without following proper sdlc with proper dod.
Better forbid them banning direct installs tbf. And while at it, do the same for iOS.
Now we need to invent a recursive acronym for this name and it would be perfect
some hacker unleashes malicious AIs to the internet, breaking it apart cause AI keeps finding vulnerabilities in everything and break things faster than humans can fix
corporates build corporate internet and the blackwall, which is AI to fight malicious AIs
Gooooood morning Night City!
AI analysis of the code flow & AI-assisted debug (placing print statements, reproducing the issue, duping the print into it) transforms some 2-3 day investigations into half a day of work.