More doom trolling as PR
Who manages the agents?
91–100 of 108 posts
Re: Who manages the agents?
#92"What should the future of humanity look like?" Some of these people have lost their damn minds. People building an agent framework that will struggle to correctly infer that my appointment at a hospital will require additional travel time when organising my calendar for me waxing lyrical about the future of the humn race is chaotic behavior. Th Wright brothers would have had no credibility discussing what ATC protoc…
The entire bubble, boom if you prefer, is built on this. Basic knowledge work can be done by robots, goodbye humans. It's serious and profound. It will come in fits and starts unless there is popular revolution against the elite criminals.
[dead]
Re: Who manages the agents?
#93Agents manage agents. The middle management in companies is one of the worst inventions ever. I think baboons have better middle management structure than us. Might as well replace all that.
> Might as well replace all that. You know, as an IC, if I get get the level of introspection you can get from building distributed systems, and that with reasoning/thinking traces from LLMs, I think I'd prefer the entire level of middle management made out of LLMs rather than humans. It'd be helpful to be able to see exactly where their logic suddenly took a skip out the window.
Same, LLMs also don't have the same biases, workplaces could become more accessible and inclusive and the managers would actually have deep knowledge about the subject.
Re: Who manages the agents?
#94Earlier quoted context omitted.
Serious question then: if you were the CEO/CTO and were employing 10,000 engineers, what would you do without middle management? Would you have 1k+ line managers each managing I'm not saying corporations execute on this flawlessly, just any time when I wonder what I would've done myself, I end up with a similar structure (maybe a bit flatter)...
I heard Microsoft has 14 layers of middle management just for the xbox team. 14 layers!! There are people whose sole purpose is to have meetings to pass information to other people, who will then have meetings about who to have meetings with.
[flagged]
Re: Who manages the agents?
#95Re: Who manages the agents?
#96Earlier quoted context omitted.
I am seeing a lot of people suffering from AI psychosis getting interested in writing a trading bot using large language models. They give them access to basic market data and then ask the model for analysis of the current market conditions. The bot talks about seeing high volume here, that being a confirmation for this, and that doing an invalidation of that etc. I looked into it after someone pitched me theirs, onl…
I found a scam online about 18 months ago that was selling access to some "AI trading platform" that was supposed to have the kind of bot you mention. I used the live chat to talk to someone and asked about demonstrable proof - like given data up through the end of last month, make picks for this month, then look at this month's data to see how it would have done. "Then this product is not for you." Right, not for an…
but wouldnt that depend on your pompts though?
Re: Who manages the agents?
#97Earlier quoted context omitted.
It should be reserved for a human because humans have a body and a body can be put in prison. A bit dramatic for effect but true.
Yes, this. For a slightly less dramatic version - you can fire a human if they consistently do the wrong thing despite being told how to do it better. Putting a big ball of matrix arithmetic on a Performance Improvement Plan makes no sense.
I don’t know, I can imagine that the equivalent of a PIP for an agent would be noticing flaws in its output and then creating additional evals, modifying the harness, upgrading the model, etc. to improve its performance. That’s not SO different from the PIP, except that the LLM doesn’t really care in the same way?
Re: Who manages the agents?
#98Maybe a useful concept to adopt here is the DRI - Directly Responsible Individual, which is apparently a term first used by Apple. The GitLab Handbook has a good definition: https://handbook.gitlab.com/handbook/people-group/directly-r... "ultimately accountable for the success or failure of a specific project, initiative, or activity" I think that role should be reserved for a human, who can then use all the agents t…
A security measure I use on my projects is never to give the AI models in production write authority at all - AI simply produces the "suggestion" in a structured format and then deterministic code validates it by making sure it conforms to pre-set constraints. This way, the blast radius of a really dumb decision is minimized. I do the same thing essentially in the codebase as well where we have architectural static analyzers that box in the agents and limit how stupid they can be with their changes.
Re: Who manages the agents?
#99Earlier quoted context omitted.
What do you think the avg execs info exposure is to what's really happening
An exec can ask an expert to explain themselves, and if they're a good CEO they can weigh the pros and cons of the arguments from their staff and make a decision. Someone that's much smarter or more skilled in a particular discipline can usually explain what's going on to someone that's not as smart as long as the gap isn't too big. However, I've found that with Fable, and some of the latest frontier models their cri…
https://arxiv.org/abs/2606.16475
Pretty sure it can dumb it down to cave man speak for any concept
Re: Who manages the agents?
#100Earlier quoted context omitted.
Yes, this. For a slightly less dramatic version - you can fire a human if they consistently do the wrong thing despite being told how to do it better. Putting a big ball of matrix arithmetic on a Performance Improvement Plan makes no sense.
I don’t know, I can imagine that the equivalent of a PIP for an agent would be noticing flaws in its output and then creating additional evals, modifying the harness, upgrading the model, etc. to improve its performance. That’s not SO different from the PIP, except that the LLM doesn’t really care in the same way?
The person making the tweaks to the harness is the one taking accountability for fixing those mistakes - they're the DRI in this scenario.