The Trillion Dollar AI Software Development Stack

**Guido Appenzeller** (0:00)
AI coding is the first really large market for AI.

**Yoko Li** (0:04)
When do we say, this is all agents? We just at the end of the value chain, we're like, does this work or not work? I click yes or no. Agents more than ever need an environment to run these things. Context engineering for both humans and agents.

**Guido Appenzeller** (0:19)
Every single part of it is getting disrupted. It's not that there's just somebody writing code, like your classical developers being disrupted, but everybody along the value chain. So, Yoko, we just launched this, I think, amazing new dev stack for the AI coding environment. And I'm really excited about this.

**Yoko Li** (0:43)
Yeah.

**Guido Appenzeller** (0:44)
And let me start with a very high-order pitch, why I think this is so incredibly exciting. I think AI coding is the first really large market for AI, right? I mean, we've seen there's a ton of investment that has flown, and the question now, to some degrees, as far as the value, right? Why are we doing all this? AI coding can create an incredible amount of value. If you think about this, we have about 30 million developers worldwide, roughly, right? Let's say each of them generates $100,000 in value. In the United States, it may be low because many of them get paid a lot more, but internationally, it might be a little high. But I think in order of magnitude, it holds. So in aggregate, the value we're creating here is about 30 million times 100,000, so $3 trillion.

**Yoko Li** (1:27)
I will argue even more because that's just developers, but then there's also people who are development and curious. They were not developers, maybe they're, I mean, design engineering, now it's a big thing. Every designer should just write code. Product managers, doc writers.

**Guido Appenzeller** (1:40)
Exactly.
There's so many of these effects. But if you just take the $3 trillion figure, that's about the GDP of France. The claim we're making here, as crazy as it sounds, is that we're saying, the entire population of the seventh or eighth, I think, largest economy on the planet that generates about as much value as a couple of start-ups that are reshaping the AI software development ecosystem plus the LLMs underneath.

**Yoko Li** (2:04)
Yeah.

**Guido Appenzeller** (2:05)
That's crazy.

**Yoko Li** (2:05)
Everything we see and touch and use nowadays are all software.

**Guido Appenzeller** (2:10)
That's right. Yeah. Software has disrupted everything in the world, and now software itself is getting massively disrupted.

**Yoko Li** (2:19)
Totally. And then what you mentioned in the blog post is really interesting, just because we now are more capable at using LLM to generate code and coding and produce software. But then as a result, it's not like there's less jobs. It's actually more and more software is being produced. Before, maybe it's a SaaS service, catering to hundreds of people's needs, to thousands of people's needs. Now you can really just vibe code things, software for one. Yeah. I do that. Exactly. You do that exactly. And then I vibe code my own email filter. I don't do that so much of using LM to reply to my email, but I have a filter where I categorize the labels, things like that.

**Guido Appenzeller** (3:00)
Only to some emails.

**Yoko Li** (3:01)
Only to some emails. So the first question becomes like, how do you think the development loop is shifting?

**Guido Appenzeller** (3:10)
I think the answer is complex. And very frankly, it's so early, I think, in this AI revolution, we don't have the full answer yet.
But we have our little stack, our little software development life cycle post AI in the blog post. And I think the biggest learning from that is probably that every single part of it is getting disrupted. It's not that there's just somebody writing code, like your classical developer is being disrupted, but everybody along the value chain is getting disrupted.

**Yoko Li** (3:37)
What's the most surprising part for you? What's the most disrupted field today in coding? And what do you think it will be? Like what AI will come after next?

**Guido Appenzeller** (3:46)
So I think we've seen the biggest growth, I think it's safe to say, in the classic sort of coding, IDE integrated coding assistance or more agentic coding assistance, right? You know, the cursors and davens and GitHub copilots and cloud codes of the world, right? I think that's where we see the most traction, where we see an incredible revenue growth. Possibly, I mean, I want to say that segment possibly has the fastest revenue growth of any startup sector we've seen in the history of startups, which is again, an incredible statement. So I think this is currently the vanguard, right? And everybody's aware of it. We're seeing, you know, billion-dollar acquires or takeover offers. So that's an incredibly vibrant sector. Now, which one is next? That's a really good question.