Scaled Cognition

This account is verified because it's an official organization on X. Learn more

@ScaledCognition

The only frontier model for CX that eliminates hallucinations.

SF, NYC, Boston scaledcognition.com Joined January 2023

69 Following 1.1K Followers 169 Posts

Latest posts 5

Pinned post

Scaled Cognition @ScaledCognition Jun 25

AI has a reliability problem. We're fixing it. 💪

Dan Roth Verified account @roth_dan Jun 25

LLMs lie. We build models that tell the truth.
Today, we're excited to announce our $100M Series A led by @vkhosla and @KhoslaVentures.
@profdanklein and I founded @ScaledCognition to solve the key challenge in AI, reliability.

View quoted post on X

Scaled Cognition @ScaledCognition 2d

“When you retrofit guardrails on, you get a much worse outcome.”

Yesterday our CTO @profdanklein sat down with @sophiadew on @mtslive to talk about why we’re building Super-Reliable Intelligence.

1:27

Scaled Cognition @ScaledCognition 3d

Tune in today at 11:55 PT / 2:55 ET to catch @profdanklein live on @MTSlive, where he’ll talk about why we are building super-reliable intelligence.

MTS @MTSlive 3d

ANTHROPIC IPO | SHORT TIMELINES | MERCOR RAISES x.com/i/broadcasts/2…
View quoted post on X

Scaled Cognition @ScaledCognition Nov 5

We’re excited to welcome Brennan Smith to the team!

As a Solutions Engineer, Brennan has firsthand experience designing and building CX agents for customers, where complex policy and scale push the limits on agent reliability. Compensating for this leads to increased latency and costs, but it shouldn’t have to.

“Approaching the problem from first principles means questioning the entire stack, including the generalized LLMs that power the experience.”

For Brennan, business conversations demand models trained specifically for verified actions and reliable behavior. That’s what drew him to APT and our approach to building Super-Reliable Intelligence.

Join Brennan and our fast-growing team building the next generation of AI agents: scaledcognition.com/careers

0:15

Scaled Cognition @ScaledCognition Nov 3

“Just think of all of the things we could do with AI if you could just trust the system, if you could trust it to do what you told it to do every single time.”

Our CTO @profdanklein explains why the big unlock for the most complex enterprise workflows is reliability, not intelligence.

theaiinnovator.com/can-ai-be-reli…

theaiinnovator.com Can AI Be Made ‘Super’ Reliable? This Professor Says Yes UC Berkeley professor Dan Klein, who’s on his third startup, says businesses need AI they can trust to always follow the rules.