Building The World's Most Wearable AI | Dan Siroker (Limitless)
An inside look at crafting an AI wearable that sold 10K units in 24 hours and Dan's #1 lesson from 15 years of entrepreneurship
Dear subscribers,
Today, I want to share an inside look at building the fastest-selling AI wearable.
Dan Siroker is the CEO of Limitless and also co-founded Optimizely. Last week, Limitless sold 10,000 AI pendants in just 24 hours to help people remember what they’ve seen, said, and heard.
In our interview, Dan shared:
An inside look at crafting the Limitless AI wearable
His #1 lesson after 15 years as an entrepreneur
Thoughts on product management and managing his time with 3 kids
Listen now on YouTube, Spotify, and Apple, or read our abridged interview below.
An inside look at crafting the Limitless AI wearable
Welcome, Dan! So you announced your new AI pendant last week. How’s that going?
We had 10,000 preorders in 24 hours which has been pretty crazy. Surprisingly, 40% of buyers are from outside the US.
Incredible. Let’s go back to the beginning — why did you start Limitless?
I started to lose my hearing in my 20s but didn’t try a hearing aid until 30. It felt like gaining a superpower when I put it in.
Ever since that moment, I’ve been on the hunt for tech that can augment human capabilities.
We’ve taken a meandering path to deliver this pendant but I’m so excited about our personalized AI vision.
Can you talk more about the meandering path that you took?
Sure, we started exploring the personalized AI idea maze back in COVID:
Our 1st idea was Scribe, an online meeting bot that took notes. But customers told us that they wanted to capture what they see, hear, and say beyond meetings.
Our 2nd idea was Rewind, a Mac app that would capture everything that you did on your computer. But people wanted access across devices with privacy.
So that’s what Limitless is. It has the privacy of a local app and the convenience of the cloud. You can use it from any device (including our pendant) with privacy.
Why did you decide to extend beyond software to build an AI wearable?
Like everything else — it starts with the customer problem and the constraints.
I would have avoided building a hardware device if I could have. But customers want an AI that can capture what they say and hear. With today’s mobile OS, you can't have an always-on microphone. As soon as you get a phone call, the microphone stops.
So we had to build hardware to deliver the customer experience that we wanted.
I love how focused the Limitless AI pendant is compared to other AI wearables. It’s a $99 device that helps you with two specific use cases - meetings and personal note-taking. How can a person improve their life with this device?
It starts with recognizing our human limitations. I didn’t even realize that I had hearing loss until I bought a hearing aid.
But in my opinion, our biggest human limitation is memory.
90% of what you experience, you forget after just one week.
We live our lives like Dory in Finding Nemo, forgetting most of what we experience. Our pendant helps augment your memory with meetings as the first use case:
It helps you prep for the meeting
It takes notes for you
It recaps the takeaways and next steps
All of this lets you be more present in a meeting instead of doing busywork.
How will the AI pendant protect the privacy of meeting attendees?
Great question. We know that Limitless needs to respect the privacy of not just the user but also the people they’re talking to. So we built:
Consent mode to keep track of who’s speaking. We only record someone’s voice if they verbally consent (e.g., “Can I record this?" "Sure, no problem.").
Confidential cloud to combine the privacy of a local app with the convenience of being in the cloud.
Dan’s #1 lesson after 15 years as an entrepreneur
You’ve said that your most important lesson is to “ship fast and hire slow.” How do you “ship fast” while keeping the quality bar high?
I still have to remind myself of these four words all the time despite being a founder for 15 years. It starts with recognizing that:
You can only have 2 out of 3 things — scope, quality, and speed.
My advice is to pick quality and speed every time. Tech people want to build all kinds of cool features that don’t matter. Instead, start by solving a narrow use case well and then expand from there. That’s why we’re focusing Limitless on meetings and personal note-taking first.
But how do you decide what to focus on and what scope to cut?
Great question. The reality is that most people don’t do these two things: