this post was submitted on 20 Feb 2026
45 points (100.0% liked)
Technology
1376 readers
112 users here now
A tech news sub for communists
founded 3 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
a guy i worked with at my last job is forming a company doing something akin to this.. he's been working on it for 6 months or so. he's a math wizard and believes he can get this to work in a way that can be mathematically proven but to be honest when he talks about it it all goes way over my head pretty quickly.
i hope you're both right because it would be great to not have to wrestle models to do things and if you can give a prompt for what you want done and the orchestrator can break it down into workable tasks and then pass those tasks out to agents to do, check, verify, in a way that is reliable it will be a game changer for sure.
the downside will be most of IT jobs will be gone pretty quickly
I think you'll still need a human in the loop because only a human can decide whether the code is doing what's intended or not. The nature of the job is going to change dramatically though. My prediction is that the focus will be on making declarative specifications that act as a contract for the LLM. There are also types of features that are very difficult to specify and verify formally. Anything dealing with side effects or external systems is a good example. We have good tools to formally prove data consistency using type systems and provers, but real world applications have to deal with outside world to do anything useful. So, it's most likely that the human will just work at a higher level actually focusing on what the application is doing in a semantic sense, while the agents handle the underlying implementation details.