Agentic coding forces you to design clean interfaces and document them well. An agent cannot read the implicit mental model shared by your engineering team, it can only read your API contracts and docstrings.
When the cost of execution drops, the value of taste, strategy, and architectural vision skyrockets. Previously, you were spending most of your cognitive budget on the micro. Now you are free to focus on the macro.
Market efficiency is a myth. Markets are made of people, who are irrational both individually and in aggregate.
Maybe if human irrationality were completely random, it might average out to a rational whole. But in practice individual irrationality is driven by globally shared irrational narratives -- almost everybody is an irrational conformist.
But if it is in the middle of a task, and it reasons a token that generates a decision to click at speed...while training another agent about loops it made. Why do you choose to think of the click automation here outside the set of decision? The click is not random. That's the whole point of choice.
"A system of agents startups are a group of agents that actually collaborate amongst each other while recognizing that there is a agent hierarchy"
"They see the full context at decision time: what inputs were gathered across systems, what policy was evaluated, what exception route was invoked, who approved, and what state was written. If you persist those traces, you get something that doesn’t exist in most enterprises today: a queryable record of how decisions were made."
"We call the accumulated structure formed by those traces a context graph"
All copied from: @JayaGup10
Don't fooll yourself by hopping across models. Stick with one and build a class experience that the model will know what to do every time you hold a conversation with it. Avoid model hoping. Stick to 5.5!
Thoughts on the origin of Inductive Bias: Humans possess a faculty that recognize order
Generalization requires a built-in intellectual tendency to detect order, analogy and class structure in experience. This order detection mechanism emerges from compression. But that is not the full story.
Let's make this safe assumption:
The model gets really good that predictive processing that it learns to dream within the human encoded intentions.
Are human inductive biases equivalent to a non-uniform metric over hypothesis space?
Safe assumption.
Hiring?
Interview:
A: Where is your personal backend?
B: huhhh?
A: Don't you have a harness?
B: OH. Gimme a sec. Clicky, let's do the loopy thing. Start with the OmegaZ skill.
@alexwg@bhalligan When people log into social media accounts, they use credentials.
Say you have a digital avatar that lives with an environment designed by you, and someone else's digital avatar wants to pay a visit and brainstorm with yours? What is the verification method?
@shuying_luo@rasbt What consititute a context file for you? If they are Souls, heartbeat, memory etc and skills sets, are their more to it. If they are just these ones, then why not create a multiagent profile where each profile is an agent with a soul, memory, etc?
@rasbt@monsieur_pickle Can you provide your view on the the difference between harness and models. Can you recommend a decent articles? Thank you so much.
Thoughts on the origin of Inductive Bias: Humans possess a faculty that recognize order
Generalization requires a built-in intellectual tendency to detect order, analogy and class structure in experience. This order detection mechanism emerges from compression. But that is not the full story.