Jensen Huang's Leather Jacket Hides a $3,000 Mistake

Jensen Huang has a net worth north of $100 billion. His company, Nvidia, briefly became the most valuable on Earth. He signs autographs like a rock star and packs arenas in Taiwan and San Jose. So here's the part nobody puts in the keynote. The man still talks about almost running out of money. Not once. Constantly. For years. In 1993, Huang co-founded Nvidia with two friends at a Denny's in San Jose. They were designing graphics chips for video games. It was a crowded, low-margin business, and they had no idea if anyone would buy what they built. By 1995, the company was nearly dead. Its first chip flopped. Sega, its one big customer, pulled out. Huang later said the company had about six months of cash left — and he meant it the way you mean it when you're staring at a mortgage payment you can't cover. Here's the move that saved them, and it's the one worth stealing. Instead of doubling down on the failing chip, Huang went back to Sega and asked to be released from the contract. Then he asked them to keep paying anyway. Sega's CEO, Shoichiro Irimajiri, agreed to a $5 million payment for work Nvidia would never finish. That cash bought Nvidia time to build the RIVA 128, the chip that actually worked. That was 1997. Nvidia went public in 1999. The rest is the AI boom you've been hearing about for two years straight. Now the household-budget part, because that's what you're really here for. Huang's story gets told as a triumph of vision. It's really a story about cash flow. He didn't win because he was smarter. He won because he kept the lights on long enough to be right. Six months of runway taught him something most people learn too late: the business that survives isn't the one with the best idea. It's the one that doesn't run out of money first. That's true for Nvidia. It's true for your checking account. Look at your own numbers this week. Not your goals. Your runway. If your income stopped tomorrow, how many months could you cover rent, groceries, insurance, and the minimum payments? If the answer is less than three, you're in the same spot Huang was in 1995 — except you don't have a Sega CEO to call. Three things that actually move the needle: One, find your real monthly burn. Add up everything that leaves your account for 30 days. Most people guess low by 20%. Two, build a one-month buffer before you build anything else. Not six. One. A single month of expenses in a separate account changes how you make decisions. You stop taking the bad client. You stop panic-selling. Three, cut one recurring charge this week. Not ten. One. The subscription you forgot about. The plan you upgraded and never use. That's your version of the Sega check — small, unglamorous, and it buys you time. Huang still wears the leather jacket. He also still tells the Denny's story every chance he gets, because he knows the ending was never guaranteed. Your ending isn't either. But runway is the one thing you can actually control. **The takeaway:** Everyone wants to talk about the vision. The vision is cheap. What kept Nvidia alive was a boring, uncomfortable conversation about how many months of cash were left — and most families never have that conversation until the bank forces it. Have it this weekend, before anyone forces you.