High Signal Podcasts Evidence ledger
Method
Browse
How the evidence works

Public evidence record

Andrew Lee

Published podcast speaker

Claims
21
Episodes
1
Shows
1
Named items
1

Books, apps, and tools

The evidenced stack.

Browse the grouped index →

service / uses

Anthropic

“In the case of Anthropic, we're using five-minute caching, and so it doesn't stick around very long.”

The Cognitive Revolution · 15 May 2026

Evidence receipt · Source ↗

Claim ledger

What Andrew said.

21 transcript-backed records

02 / prediction

sophistication of these harnesses is going to get just 10 times as complex. But I think there's going to be some pretty major breakthroughs here that increase the capabilities of these things pretty substantially in the way that we handle memories and the way that we handle oversight and control and the way we connect to other tools.

“sophistication of these harnesses is going to get just 10 times as complex. But I think there's going to be some pretty major breakthroughs here that increase the capabilities of these things pretty substantially in the way that we handle memories and the way that we handle oversight and control and the way we connect to other tools.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

03 / belief

I think the value of being a system of record in a world where you have agents goes down a lot because like moving data around between systems suddenly gets a lot easier.

“I think the value of being a system of record in a world where you have agents goes down a lot because like moving data around between systems suddenly gets a lot easier.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

04 / belief

I agree with that, and I think that trend is going to continue. But I also think the effects are multiplicative, and they're orthogonal disciplines, and there's no reason not to take the best model and put in the best harness, and I think we should.

“I agree with that, and I think that trend is going to continue. But I also think the effects are multiplicative, and they're orthogonal disciplines, and there's no reason not to take the best model and put in the best harness, and I think we should.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

05 / belief

I think the harnesses are also kind of converging in terms of capabilities, and largely that's because it turns out the best harness is just through low-level primitives.

“I think the harnesses are also kind of converging in terms of capabilities, and largely that's because it turns out the best harness is just through low-level primitives.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

06 / belief

There are some changes that we can make to do a lot of cash optimization across agents and potentially even across organizations and users. And I don't want to get into the specifics of that because that's still upcoming, but I think there's a lot of potential actually to save money across agents as well.

“There are some changes that we can make to do a lot of cash optimization across agents and potentially even across organizations and users. And I don't want to get into the specifics of that because that's still upcoming, but I think there's a lot of potential actually to save money across agents as well.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

07 / belief

We have lots of ideas how to improve this, but the basic approach of this decreasing fidelity as you go back and these bucketed cache-aware chunks, I think is the right approach.

“We have lots of ideas how to improve this, but the basic approach of this decreasing fidelity as you go back and these bucketed cache-aware chunks, I think is the right approach.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

09 / belief

There's going to be the horizontal platforms, of which I think there'll be a very few winners because people don't want to have to maintain context and connection across a bunch of platforms.

“There's going to be the horizontal platforms, of which I think there'll be a very few winners because people don't want to have to maintain context and connection across a bunch of platforms.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

10 / belief

I think ideally as little as possible because we want to support a lot of models and like it's hard to have a thing that, you know, It's hard to maintain across both balls, but we also want to have the ability for these agents to switch between models.

“I think ideally as little as possible because we want to support a lot of models and like it's hard to have a thing that, you know, It's hard to maintain across both balls, but we also want to have the ability for these agents to switch between models.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

11 / belief

The naive way to do this is to like load that data through an API, feed it into the LLM, have the LLM then like call some tools to put it somewhere else. And basically, when you do that every time, you're putting it through language model context and trusting it to not hallucinate and reproduce that data, which I think the models get better at over time, but it's very hard to have a lot of confidence there.

“The naive way to do this is to like load that data through an API, feed it into the LLM, have the LLM then like call some tools to put it somewhere else. And basically, when you do that every time, you're putting it through language model context and trusting it to not hallucinate and reproduce that data, which I think the models get better at over time, but it's very hard to have a lot of confidence there.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

12 / evaluation

The number one reason that they do that is because they already have a max plan, and they don't want to have to spend additional money on tasks.

“The number one reason that they do that is because they already have a max plan, and they don't want to have to spend additional money on tasks.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

13 / evaluation

And usually the answer is no. And I think the ones that we've, like Kimi, DeepSeek, and Google, and plus obviously OpenAI are the ones where like, Okay, actually this is pretty close to the frontier, so it's worth doing.

“And usually the answer is no. And I think the ones that we've, like Kimi, DeepSeek, and Google, and plus obviously OpenAI are the ones where like, Okay, actually this is pretty close to the frontier, so it's worth doing.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

14 / evaluation

We've also rethought the way our integrations work. So this is something I think for a product experience that maybe doesn't look terribly different to folks, but the basic architecture of how we plug in other systems to the Tasklet agents has been completely rethought, basically to allow the agent to have sort of more control and management of like, those connections.

“We've also rethought the way our integrations work. So this is something I think for a product experience that maybe doesn't look terribly different to folks, but the basic architecture of how we plug in other systems to the Tasklet agents has been completely rethought, basically to allow the agent to have sort of more control and management of like, those connections.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

15 / evaluation

The reason we started on Anthropic and I've been so Anthropic focused for so long was just that the basic core of our agent, like the ability for it to navigate through a discovery process of connections and activate the right tool in your agents and then manage its context the way we manage its context, just the core inner workings requires kind of a base level of intelligence that the other models just couldn't do.

“The reason we started on Anthropic and I've been so Anthropic focused for so long was just that the basic core of our agent, like the ability for it to navigate through a discovery process of connections and activate the right tool in your agents and then manage its context the way we manage its context, just the core inner workings requires kind of a base level of intelligence that the other models just couldn't do.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

16 / evaluation

The reason our context manager is built the way it is is because you want to have triggers as sort of regular messages into the agent, which means these agents, if you have an agent that's running a trigger every time you get an e-mail, that agent might fire 10,000 times this year.

“The reason our context manager is built the way it is is because you want to have triggers as sort of regular messages into the agent, which means these agents, if you have an agent that's running a trigger every time you get an e-mail, that agent might fire 10,000 times this year.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

17 / preference

Then we built the thing we launched in October, and the feedback from people was like, Hey, We don't wanna have one tool for workflow automation and another tool for doing our day-to-day work because we want them all to have the same context.

“Then we built the thing we launched in October, and the feedback from people was like, Hey, We don't wanna have one tool for workflow automation and another tool for doing our day-to-day work because we want them all to have the same context.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

18 / evaluation

from a competitive position, like, you know, is there concern here that like OpenAI isn't going to like own the user relationship and, you know, if they already have an OpenAI account, why do they have an account with us? I think we are maybe a little more concerned now than we used to be.

“from a competitive position, like, you know, is there concern here that like OpenAI isn't going to like own the user relationship and, you know, if they already have an OpenAI account, why do they have an account with us? I think we are maybe a little more concerned now than we used to be.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

19 / prediction

You saw the progress Codex made over six months, and if they put that level of effort into tuning the models around these types of agenda use cases, I think that'll be It'll be huge.

“You saw the progress Codex made over six months, and if they put that level of effort into tuning the models around these types of agenda use cases, I think that'll be It'll be huge.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

20 / evaluation

Yeah, so caching has actually become a much bigger deal because now that the real context is in the file system, there's just a lot more tool calls that need to be done to do the basic operations of the agent because you're loading in a bunch of files and stuff.

“Yeah, so caching has actually become a much bigger deal because now that the real context is in the file system, there's just a lot more tool calls that need to be done to do the basic operations of the agent because you're loading in a bunch of files and stuff.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution

21 / commitment

Marc Andreessen had a thing about this, and I think people figured this out, but we made the switch kind of in November where we said, really, okay, what we need is a file system.

“Marc Andreessen had a thing about this, and I think people figured this out, but we made the switch kind of in November where we said, really, okay, what we need is a file system.”
Speaker
Andrew Lee
Publisher
The Cognitive Revolution
Search evidence