Chapter 11What Is the Atlas Reasoning Engine?
Reasoning engines combine models, data, business logic, events, and workflows into unified cognitive architectures,” says Phil Mui, PhD, SVP of Technology, Head of Products and Engineering, Salesforce AI Research and Agentforce – and he probably has the longest title in the company.
Phil is very smart, of course, but some of us might wonder what a “unified cognitive architecture” is and how it helps us build our agents in Agentforce.
The architecture he's referring to is the Atlas Reasoning Engine. It's the “brain” of Agentforce, and Phil guides it. And he frequently points out that its structure is closely related to the way we think.
Yes, we – as in human beings.
More than a decade ago, Daniel Kahneman published an extraordinarily influential book called Thinking Fast and Slow, about the human brain. In it, through a tumult of vivid and humbling examples, he described the brain as having a fast-twitch reactive side and a slower more reasoned side. He called these different modes of thought System 1 and System 2.1
It's the difference between looking at your wife and just knowing you're in trouble (System 1) and settling down in your La-Z Boy to try to figure out why (System 2). System 1 thinking is what lets us react to stimuli – real or imagined, physical or social – and happens automatically. This is what we use when we flinch or answer easy questions (“What's a Taylor Swift?”). System 2 is harder and requires conscious thought. ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access