The Unsexiest Problem in AI
Wang: The problem we originally were solving — building really high-quality datasets — was something that most machine learning teams knew was very important. Most people agreed it was very important. But it wasn't necessarily the sexiest problem that every AI scientist wanted to spend their days and nights working on.
Collecting images, labeling them, making sure that was done very high quality. It was a very operational problem at its core.
Schlep Blindness
Wang: There's one article that was pretty seminal for me early on — an essay by Paul Graham called Schlep Blindness. The idea was that most people avoid thinking about the really hairy, ugly, difficult, annoying problems that exist in the world. But they're really, really important.
He actually uses Stripe as one of the examples. These problems are everywhere — the ugly, hairy problems that everyone knows are important but aren't sexy to work on. If you can identify what those problems are, they generally make really exciting startup ideas.
Most people avoid thinking about the really hairy, ugly, difficult, annoying problems. But they're really, really important.
Velocity Won Them Over
Wang: Even very early on, our whole team was super scrappy. A lot of customers that we worked with saw our velocity — how fast we were moving — and thought to themselves, these guys, even if they don't have the perfect product today, they're going to get to a product we can rely on really, really quickly.
Even if they don't have the perfect product today, they're going to get to a product we can rely on really quickly.