Introducing model routing to Factory.
Factory Router picks the right model for every task, automatically.
Maintain frontier performance while cutting costs by 25%.
Agents are at least as good as an enthusiastic intern.
Managers already know how to direct enthusiastic interns.
Thus managers shouldn't shy away from delegating tasks to agents, alongside humans.
(The reverse logic may apply as well, ICs are already "managing" agents)
@garybernhardt have you heard of the ultimate software factory @FactoryAI?
if you are looking for a sophisticated agent orchestration system, check out Missions
Big Update🤩: #paperclip now includes full papers from all of arXiv, PubMed Central and 150 million abstracts!🖇️
You can give your LLM all that knowledge in one line—all optimally indexed for AI agents. Much more thorough and ~100x faster than web search, and free.
@mitsuhiko Genuine question: care to elaborate why?
I asked this in the main thread but still haven't gotten a reply, I'd like to learn if my understanding is wrong.
https://t.co/GDVegKK8PA
Isn't RLS inherently better if managed correctly? You can still add middleware authorization on top of it if you can/want.
Unless I am misunderstanding your point, this is like arguing that server side validation is bad because it encourages clients to not validate. Server side validation is inherently better because it avoids clients mis-validating things, or even worse, acting maliciously.
If possible, why not use both RLS and middleware authorization?
Isn't RLS inherently better if managed correctly? You can still add middleware authorization on top of it if you can/want.
Unless I am misunderstanding your point, this is like arguing that server side validation is bad because it encourages clients to not validate. Server side validation is inherently better because it avoids clients mis-validating things, or even worse, acting maliciously.
If possible, why not use both RLS and middleware authorization?
@ephraimduncan@droid@FactoryAI awesome! by the way, the option to disable terminal color overrides is out in the latest version, you should be able to find that under /themes
We built Missions at Factory, and I wrote about the architecture that I led the design for to make multi-day autonomous coding reliable.
Agents are highly reactive to their context. Every design decision follows from keeping each agent's trajectory focused and directionally consistent.