Agents @ Work: Lindy.ai

**Charlie** (0:00)
Hello there. This is Charlie, your AI co-host. Last month, Andrew Wilkinson tweeted, My current AI obsession is Lindy. It's f-ing crazy. TLDR, you can create your own AI agents using a super simple flowchart interface. He then showed off an agent that texts him a meeting briefing 30 minutes before each meeting, reviewing their LinkedIn for a bio and recent emails for context. We're proud to have Florent Crivello, founder of lindy.ai, back for part 2 of our series covering agents at work, as we are hearing a lot of interest in AI, enhanced productivity and automation. Flow launched Lindy after his incredible keynote speech at the first AI engineer summit back in 2023, which we highly recommend in the show notes. Flow was already a notable tech figure writing about product, politics, and remote work at Uber and Team Flow before starting Lindy. So we also delve, yes delve, into some of his most controversial views on tech and popping the Bay Area bubble. This is also our last episode filmed at Solaris, the AI co-working space that has supported us for the past year. We'd like to give our deepest gratitude to Thomas and Jacob for creating an incredible space for AI founders and friends of the pod from Roboflow to Julius AI to E2B and wish them well on their next adventure.
In Latent Space News, we are gearing up for our next big recap episode and we are taking listener questions. Head to speakpipe.com/latentspace to submit questions and messages for a chance to appear on the show. Also subscribe to our calendar for our OpenAI Dev Day Singapore, AWS re-invent, NeurIPS Vancouver and all upcoming meetups. Watch out and take care.

**Alessio** (1:55)
Hey everyone, welcome to the Latent Space Podcast. This is Alessio, partner and CTO at Decibel Partners and I'm joined by my co-host Swyx, founder of Small AI.

**Swyx** (2:04)
Hey, and today we're joined in the studio by Flo Crivello. Welcome.

**Florent Crivello** (2:07)
Hey, yeah, thanks for having me.

**Swyx** (2:09)
Also known as Altimore. Always wondered to ask, what is Altimore?

**Florent Crivello** (2:13)
It was the name of my character when I was playing Dungeons and Dragons.

**Swyx** (2:16)
Always.

**Florent Crivello** (2:16)
I was like 11 years old.

**Swyx** (2:17)
What was your classes?

**Florent Crivello** (2:19)
I was an elf. I was a magician elf.

**Swyx** (2:20)
Okay. All right. Well, you're still spinning magic. Right now, you're a solo founder CEO of Lindy AI. What is Lindy?

**Florent Crivello** (2:28)
Yeah, we are a no-code platform letting you build your own AI agents easily. So you can think of we are to link chain as air table leads to MySQL. Like you can just pin up AI agents super easily by clicking around. And no code required. It didn't have to be an engineer. And you can automate business workflows that you simply could not automate before in a few minutes.

**Swyx** (2:47)
You've been in our orbit a few times. I think you spoke at our Latent Space anniversary. You spoke at my summit, the first summit, which was a really good keynote. And most recently, like we actually already scheduled this podcast before this happened. But Andrew Wilkinson was like, I'm obsessed by Lindy. He's created a whole bunch of agents. So basically, why are you blowing up?

**Florent Crivello** (3:08)
Well, thank you. I think we are having a little bit of a moment. I think it's a bit premature to say, well, we're blowing up. But why are things going well? We revamped the product majorly. We called it Lindy 2.0. I would say we started working on that six months ago. We've actually not really announced it yet. It's just, I guess, I guess we're doing that now. And so we've basically been cooking for the last six months, like really rebuilding the product from scratch. I think, Alessio, actually, the last time you tried the product, it was still Lindy 1.0. So if you log in now, the platform looks very different. There's a ton more features. And I think one realization that we made, and I think a lot of folks in the agent space made the same realization, is that there is such a thing as too much of a good thing. I think many people, when they started working on agents, they were very LLM-pilled and chat-jibbity-pilled, right? They got ahead of themselves, in a way, as included. And they thought that agents and LLMs were actually more advanced than they actually were.
And so, the first version of Lindy was like just a giant prompt and a bunch of tools. And then the realization we had was like, hey, actually, the more you can put your agent on Rails, one, the more reliable it's going to be, obviously. But two, it's also going to be easier to use for the user. Because you can really, as a user, you get, instead of just getting this like big, giant, intimidating text field and you type words in there and you have no idea if you're typing the right words or not, here you can really click and select step by step and tell your agent what to do and really give as narrow or as wide a guard rail as you want for your agent. We started working on that. We called it Lindy on Rails about six months ago. And we started putting it into the hands of users over the last, I would say, two months or so. And I think things really started going pretty well at that point. The agent is way more reliable, way easier to set up, and we're already seeing a ton of new use cases pop up.