Aw.

You know, we talk every week.

And then, uh, two of them started to go through a pretty gnarly divorce.

Um, and my parents got divorced too, so, you know, I was like, "Divorce? Hmm," like, no one seems to be talking about divorce in AI.

What...

Why? Like...

And so I just started working on that, um, and was, of course, empowered by Thor's hammer, you know, which is the large language models, and, and I started to build it.

It's a good one.

So I, I keep hearing this, that Devon is, is almost, you know, it's the most powerful coding agent and, you know, can you just for folks who haven't heard of Devon or, or who haven't played with it, like why is that?

What is going on? So, um, so about two years ago, maybe two and a half years ago, Devon came out, and they pitched it as this idea of like, "It's a software engineer," right? And honestly, it didn't work very well, right? The, the models, uh, were just not good enough.

And no matter how good the harness was, the models just weren't there.

Um, and then fast-forward to, you know, 2026, um, the models are definitely good enough.

And when you have a good harness that is cloud-based, um, it's really good.

So think of it as, you've got your choices, right? So if you're using agents to build things, you're probably using Codex, you know, from OpenAI or using Claude Code from Anthropic, or you're using one of the indie, I call them the, the indie, uh, agent harnesses.

You've got Amp, you've got Cursor, which is not independent anymore, but sorta, uh, owned, owned by Elon.

Mm-hmm.

So can you tell us where was the problem that you saw?

Yes.

So I've been in the tech space, like I said, for a long time.

Uh, built and sold, um, three tech companies, and after my last one was acquired, uh, I was looking for the next thing to build.

And, uh, I am best friends with my three sisters, so I'm- I've got a sibling tattoo somewhere there.

Um, and we, we talk every week and, um, about three years ago, two of them started to go through divorce.

Uh, my parents had gotten divorced when I was a kid.

Yeah

And you wrote this awesome ar- article, maybe we can use that as a framing of walking through your setup.

So like Peter said, the truth is, is that this stuff is still really hard to use, right? So the...

A- anyone who tells you that they're, you know, waving a wand and getting OpenClaw, you know, to change their entire life and their entire business is not tr- being truthful to you.

[chuckles] So basically, you know, I, I've been fortunate to run a number of companies.

I've, I've had several real executive assistants that are humans, uh, in the past, and my goal was to turn R2, my OpenClaw, into an amazing executive assistant or chief of staff.

Um, so I started working on that.

And I built a system called O- uh, called, uh, Claw Chief.

Like what do you use to plan? What do you use to ship?

I don't want to.

Um, uh, a, a very real example, a, a very real example is this, right? Anthropic is terrible at dev rel.

They are the worst company I've ever seen do developer relations.

They just are the worst, right? And then you, you, you look at Codex.

So the Codex team in OpenAI, they are one of the best.

Their dev rel team is killer, right? But the truth is, people, as Alex said, what are we still using to do our work? And it's Opus 4.6.

So it's the model.

Mm

So that's what a good structure would look like there.

Yes.

Yeah.

Engage your brain.

You know, don't outsource your thinking.

That's the primary difference between agentic engineering and vibe coding.

I think the other thing is people should really look into concepts like compound engineering, which Karen Clarkson from Every introduced.

So I just want to, I think, differentiate between, at least from your working experiences, where mcps can actually be helpful especially from the data context that we're talking about as opposed to let's say these kind of agents directly working on our markdown files or your data wherever it is like is there any sort of a difference in perspective or output that you see

Show us what you're working on.

Okay, so essentially, I'm, I'm building a startup.

Just closed my seed today, very exciting.

And all of us that are running companies now are trying to orchestrate agents, right? Teams of agents, right? You can now do with yourself plus 10 agents what you used to be able to do with almost 100 people, right? So I've been trying to orchestrate this stuff, right? So I built an open source tool to do that.

It's called Antfarm.

It's completely free.

Everybody should try it, and you just go to ant, uh, antfarm.cool. And the way it works is, is pretty simple.

[laughing]

Let's do it.

Okay.

So, um, a friend of mine named Geoff Huntley thought up this idea called RALPH.

Um, and Geoff is super creative, um, and he launched this a while ago.

And the idea is really simple, um, but it... that's why it's so good.

Um, so yesterday, I thought, "You know? I'm just gonna post about this and give people a breakdown of what is this thing, how does it work." And, you know, you never quite know when the algo is gonna hit on X.

Um, sometimes it does, sometimes it doesn't, and this went bonkers, right? So you can see we're over 700,000 views.

And for people who don't know what Bash is, it's, it's just a way for you-