alt Hacker News> Since ChatGPT chats cannot spawn or control subchats/subganets so I had to be a bit creative in order to do multi-agent development and orchestration.
Multi-agent is mostly dead. One big agent loop running GPT6.1 absolutely dominates any weird ideas like this now. I still use sub agents for some tasks where I want to compress the action space and context. For example, testing a workflow that requires 100+ tool interactions for each logical test. Without a subagent it's really hard to guarantee a test makes it E2E. I have some very "persuasive" methods for long horizon tests that would break the main loop's dynamics.
We may not feel so compelled to exploit the system if we could figure out how to not be so wasteful. Multi agent schemes can bloat token usage by orders of magnitude. They force us to consider consuming less ideal tokens from contexts and vendors with higher variance.
HN loses its marbles when I suggest it's economically feasible to prepay for API tokens and use these as the exclusive means of performing AI work. I get done with a thousand tokens what my peers cannot do with a million. It doesn't matter how many hammers or workers you have if you don't have the aesthetic preferences and domain expertise to employ them correctly.
Replies
> Multi-agent is mostly dead
Wow that's a huge take. Seems like you're on your own with this one.. everyone I know is scaling up the number of agents they're able to keep fed with good, useful work.
OpenAI's view of multi-agent is something like: you get the same work faster, but it does not scale linearly.
Only you can decide if that speed is worth the cost.
For me Opus 5.5 will frequently make several thousand LoC where 10 would suffice. One time I got fed up and told it to repro the "bugs" it was solving for with puppeteer, and lo and behold, it could not do it overnight. Hard to believe that people let this run wild on their codebase, if they do, they are making a huge untenable mess.