It was 2006. The tech world had its rules clearly defined: CPUs ran computers, GPUs made games look pretty. That was the natural order, and everyone accepted it.

Everyone except Jensen Huang.

In NVIDIA's conference rooms, Huang was pitching something that made his board members shift uncomfortably in their seats. He wanted to pour millions into developing CUDA - a platform that would let graphics cards do general computing. To Silicon Valley's elite, this was like suggesting a Ferrari should also plow fields.

"Graphics cards are for gamers," they said. "Stick to what you know."

But Huang had noticed something others hadn't. While CPUs processed tasks one by one, like a single-file line, GPUs could handle thousands of simple tasks simultaneously, like a massive orchestra playing in perfect harmony. He believed this parallel processing power could revolutionize everything from scientific research to artificial intelligence.

The market didn't care. When NVIDIA released CUDA in beta, it was met with a collective shrug from investors. The stock price languished. Competitors mocked the strategy. For six long years, CUDA was NVIDIA's expensive pet project - burning cash while generating virtually no revenue.

Then came 2012. At the ImageNet competition, a watershed moment in computer science, researchers demonstrated that GPUs running CUDA could train neural networks faster than anyone imagined possible. Suddenly, the "gaming chip company" held the keys to the AI revolution.

What followed was one of the most dramatic reversals in tech history. The same technology that Wall Street had dismissed as a costly distraction became the foundation for ChatGPT, autonomous vehicles, and virtually every AI breakthrough of the past decade. NVIDIA's stock price increased tenfold. The company that was once 30 days from bankruptcy became the first to reach a $4 trillion market cap.

But perhaps the most remarkable aspect of Huang's gamble wasn't just the vision - it was the persistence. While other tech CEOs chase quarterly earnings, Huang held firm for six years of doubt, criticism, and financial pressure. Seven of his direct reports have been with him for over 20 years, a testament to the loyalty his conviction inspires.

Today, when asked about those early days of CUDA, Huang often recalls his time waiting tables before founding NVIDIA. It taught him something crucial about success in Silicon Valley and life.