The Serial Exit
Most founders dream of one exit. Dave Grannan has three. Vlingo (acquired by Nuance for $225M, whose technology powered the first version of Siri), Maluuba (acquired by Microsoft for $160M), and Light (acquired by John Deere).
Han Shu co-built the speech recognition behind Siri, then co-founded Wyth (acquired by Airbnb), then led ML/AI at both Airbnb and DoorDash. He holds 25 patents and an MIT PhD.
Combined, they've been responsible for $385M+ in acquisitions.
What Marr Labs Actually Does
They're building AI voice agents that are indistinguishable from humans. Not the robotic "Press 1 for billing" experience. Actual conversations that handle scheduling, sales calls, and customer support in real time.
The thesis: the $500B call center industry runs on human labor that follows scripts. An AI that can handle natural conversation at human quality makes the entire industry ripe for replacement.
Why Call Centers Should Pay Attention
The global call center market is worth $500B+. Turnover is 30-45% annually. Training costs are massive. Quality is inconsistent. And customers hate the experience.
When the people who literally built the voice behind Siri tell you they can replace human call center agents, the industry should listen.
$385M in exits. 25 patents. An MIT PhD. These aren't first-time founders chasing a trend. These are the people who built the technology that made voice AI possible, now deploying it where it matters most.
The Breakout Pattern
Serial founders don't go through YC unless they see something massive. Grannan and Shu have already proven they can build and exit. This time, they're going after the biggest target yet.
That's the breakout.
Your domain expertise is the moat.
Explore 50 startup ideas for engineers who refuse to compete for shrinking jobs.
Browse 50 Ideas