Jensen Huang

1. 1984–Microprocessor designer at AMD.
2. Late 1980sDirector of CoreWare at LSI Logic.
3. 1993Co-founded Nvidia with Chris Malachowsky and Curtis Priem.
4. 1999Nvidia IPO; released the GeForce 256.
5. 2006Launched CUDA, opening the GPU to general-purpose computing.
6. 2023–Nvidia became central to AI infrastructure demand.

• Long-term thinking

  Nvidia funded CUDA and general-purpose GPU computing for roughly a decade before there was a large market for it, absorbing margin pressure that the company defended publicly to investors throughout.

• Network effects

  The CUDA developer ecosystem is the publicly stated source of Nvidia's durable advantage — more developers and libraries make the platform more valuable to each new one.

• Systems thinking

  Nvidia sells chips, interconnect, systems and software as one accelerated-computing stack rather than as discrete components.

• Talent leverage

  Runs an unusually flat organisation with roughly 60 direct reports and no standing one-to-one meetings, publicly justified as keeping information unfiltered across the company.

• First principles

  The company's founding bet was that a fundamentally different, parallel processor architecture was required for graphics, rather than incrementally faster general-purpose CPUs.

Each model is listed only where a specific piece of public evidence supports it. The evidence line is what you should judge — the label is just an index into it.