Description
Every great corporate biography faces a choice that most refuse to acknowledge: whether to explain a company’s success or to worship it. Tae Kim’s The Nvidia Way announces itself as the former—a three-decade history built from over a hundred interviews, including with Jensen Huang himself—but the gravitational pull of $3.3 trillion in market capitalization bends the entire project toward the latter. What results is an absorbing, meticulously reported chronicle of one of the most consequential companies in modern technology, wrapped inside a hagiography that never quite finds the nerve to ask its hardest questions.
Kim, a former equity analyst turned journalist who openly admits that Nvidia was his “first big winner” as a Wall Street investor, traces the company from its founding in a Denny’s booth in 1993 through near-death chip disasters, its audacious bet on CUDA, and its current reign as the indispensable supplier of AI infrastructure. The early chapters are the book’s strongest. Curtis Priem, Chris Malachowsky, and Huang—three engineers who couldn’t even operate a consumer PC when they set out to dominate the PC graphics market—have genuine comic energy. Priem’s attempt to name the company produces rejected candidates including “Huaprimal” and “Malhuapri,” both of which sound like prescription medications. And the NV1 disaster delivers the kind of existential corporate dread that reminds you most companies die young.
Where the book thickens is in its portrait of Huang’s management philosophy. Kim renders in fine detail a corporate structure that is, on its own terms, fascinating: sixty-plus direct reports, no one-on-one meetings, no five-year plans, mandatory “Top 5” emails from every employee. Huang reads a hundred of these daily. On Sundays, accompanied by Highland Park scotch. The operating metaphor is the whiteboard—possibility and ephemerality, ideas sketched and erased. It’s compelling organizational thinking, and I spent a decade consulting on exactly these problems of internal alignment and institutional sclerosis. The flat structure, the “Pilot in Command” system, the fluid reallocation of talent across projects—these are genuine innovations in how a hardware company can avoid bureaucratic paralysis.
But Kim presents Huang’s methods with a credulity that borders on incurious. The chapter titled “Tortured into Greatness”—Huang’s own phrase—describes a CEO who publicly humiliates executives at all-hands meetings, has a cameraman zoom in on a project manager’s face while berating him, and tells his staff, “I look in the mirror every morning and say, ‘you suck.'” Kim frames this as directness, as learning opportunity, as the price of excellence. Among all the former employees Kim interviewed, he writes, “it was hard to find a dissenter.” I spent enough years inside organizations to know that when you can’t find dissenters, you haven’t looked hard enough—or the culture has made dissent too expensive to voice. The people who left bruised don’t usually agree to be quoted by name in the founder’s authorized biography. Mike Rayfield, the Tegra project manager who endured that public dressing-down, responded to Kim’s inquiry with a smiley-face emoji. What else would he say?
Kim does raise the obvious succession problem. He acknowledges that Nvidia is “almost completely dependent” on Huang, its “single point of failure.” But the observation arrives in the Conclusion and evaporates almost immediately. There’s no serious investigation of what the organizational structure looks like without the singular figure who designed it to amplify his own cognition. A flat hierarchy with sixty direct reports and no COO isn’t a system—it’s a personality. Remove the personality and you get sixty people with no one to report to.
The CUDA years provide the book’s most interesting material, and Kim narrates the bet well: $475 million sunk into a computing architecture that cratered Nvidia’s gross margins from 45 percent to 35 percent while the 2008 financial crisis destroyed demand. Huang’s refusal to capitulate to Wall Street’s short-term pressure is, in hindsight, the pivot on which the entire AI economy now turns. Kim captures a telling scene in which analysts pepper Huang with margin questions at a parking-lot lunch until one finally asks him something real—about photo editing on a Mac—and Huang’s eyes light up. That moment, where the CEO’s frustration with financial myopia meets his appetite for the technical particular, is the closest the book gets to an unguarded portrait.
Yet “sheer will” is the thesis Kim lands on, and it’s the weakest framing available. Priem himself offers a far more destabilizing observation: “Nvidia would have been a failure if NV1 had not failed.” The company’s success was contingent on its early catastrophes in ways that resist any narrative of willpower alone. Malachowsky almost went to medical school. Priem nearly accepted an offer to build Java accelerators at Sun. The entire edifice rests on accidents that Kim dutifully records but refuses to let disturb the clean arc of inevitability he’s constructing. Huang’s own discomfort with his early years—”I don’t love talking about our past”—should have been a thread Kim pulled harder, not a line he politely dropped.
The book left me thinking about a different question than the one Kim posed. Not whether Nvidia’s culture can survive Jensen Huang, but whether a culture that demands sixty-hour weeks as a “bare minimum” and frames public embarrassment as a gift deserves the uncritical admiration Kim extends it. Somewhere between the scotch-soaked Sunday emails and the cameraman zooming in on a sweating manager’s face, there’s a story about the human cost of building the world’s most valuable company. Kim walked right up to it and chose not to tell it.
If you’d like to read the full book in EPUB or MOBI format, feel free to send me an email—I’d be happy to share a free copy with you. Please reach me at: thenovaleaf@gmail.com
Reviews
There are no reviews yet.