Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Munder Difflin – Agent harness to run an office of your clones

munderdiffl.in

Post reply on HN
Print view View original

121–130 of 162 posts

Re: Munder Difflin – Agent harness to run an office of your clones

#121

Post by mrieck »

joshstrange wrote: post #63

Ok, I've been running it for a couple hours and below are my thoughts. Please note that I do find it fascinating even if most of what I'm about to say is complaining about the parts I like less. - Pipelines, not agents. Roles, not agents. I really don't like the idea of defined agents with their own prompt. I want to define roles and spin up N agents with that role. Furthermore I want pipelines "Plan -> Review Plan -…

Herdr has an API btw - I am using it to build my own agent orchestrator.

I setup a Telegram bot that relays messages to the agent orchestrator that then decides "this agent needs this plugin, this working directory for context" etc to relay the work I asked. I guess it's similar to the God orchestrator.

For pure coding tasks I just want a TUI, but for non-coding work (marketing, content gen, research) an orchestrator has been great.

Re: Munder Difflin – Agent harness to run an office of your clones

#122

Post by smrtinsert »

Is there something like that but simply a GUI web wrapper around subscriptions and memory layer? I want a web ui. I don't think in terms of these personality types though, my tasks are dynamic. For example, claude can run N agents in one window right now and be accurate about it, so I scope agent by project level.

Re: Munder Difflin – Agent harness to run an office of your clones

#124

Post by luciana1u »

fHr wrote: post #24
luciana1u wrote: post #4

an office of your clones. next they'll add a clone HR department to handle the clone performance reviews, and a clone IT guy who's also a clone and keeps filing tickets against himself.

Can't be worse then the preformative just-have-to-be-done newage bs HR reviews in modern big corpo landscape.

ok but at least the clone files its review on time instead of chasing three weeks of calendar invites to get to a meeting about the meeting

Re: Munder Difflin – Agent harness to run an office of your clones

#125

Post by ShinyLeftPad »

mrshadowgoose wrote: post #111
Aurornis wrote: post #84

I do like the embrace of The Office as a theme because it so accurately represents the dysfunction of all of the agent swarms I’ve seen: Different personalities pursuing their own little goals that are all competing with each other in subtle ways that eventually leads a funny collapse of the outcome you wanted. There’s no denying that LLMs are getting better by the month, but the current wave of LLM office and person…

In areas of applicability where GenAI to now considered to be good/great, the pattern I've observed is a frighteningly rapid progression from "abjectly terrible", to "middling and mostly more trouble than it's worth" and then "good/great". Agentic coding and image generation are probably the best examples of this. There's no guarantee that this trend will be seen with virtual assistants, but if it does, we're on the…

Nothing can go wrong if we reduce direct human interaction even further and pay some megacorporations to MITM all of it.

Re: Munder Difflin – Agent harness to run an office of your clones

#127

Post by Dylan16807 »

qarl2 wrote: post #49
xp84 wrote: post #47

Earlier quoted context omitted.

I don’t think GP is being unreasonable. Half of this webpage kinda sounds like parody and the other half serious. It’s valid feedback.

did i say he was being unreasonable? i asked him why he's pigeon-holing it. and you, too? don't hackers enjoy a little bit of playful ambiguity in their lives? must it be all serious all the time? sheesh.

If you want to over-analyze people's reactions, sure I'll help:

They're pigeon-holing because it can't be both. This isn't a situation where there's real overlap.

Ambiguity is a lot more fun when there is substantial overlap, and on top of that ambiguity for the question of "is this useful" is bad ambiguity much more often than it's fun ambiguity.

Re: Munder Difflin – Agent harness to run an office of your clones

#128

Post by Forgeties79 »

mrshadowgoose wrote: post #111
Aurornis wrote: post #84

I do like the embrace of The Office as a theme because it so accurately represents the dysfunction of all of the agent swarms I’ve seen: Different personalities pursuing their own little goals that are all competing with each other in subtle ways that eventually leads a funny collapse of the outcome you wanted. There’s no denying that LLMs are getting better by the month, but the current wave of LLM office and person…

In areas of applicability where GenAI to now considered to be good/great, the pattern I've observed is a frighteningly rapid progression from "abjectly terrible", to "middling and mostly more trouble than it's worth" and then "good/great". Agentic coding and image generation are probably the best examples of this. There's no guarantee that this trend will be seen with virtual assistants, but if it does, we're on the…

That future does not sound fun, let alone incredible.

Re: Munder Difflin – Agent harness to run an office of your clones

#130

Post by jonbaer »

chaicodes wrote: post #104
JKCalhoun wrote: post #58

Earlier quoted context omitted.

Does it work with local LLMs?

Yes, it supports opencode and qwen clis, both of them can be configured to work with local LLMs.

Why does it build Nemotron 3 Ultra on startup if Opencode was selected? Can I do Opencode to Ollama (local) or Ollama directly @ the moment?
Post reply on HN