LEADER DEEP DIVE 🕵🏻

Jensen Huang - Co-Founder, President & CEO at Nvidia

Jen-Hsun "Jensen" Huang (Chinese: 黃仁勳; pinyin: Huáng Rénxūn; Pe̍h-ōe-jī: N̂g Jîn-hun; born February 17, 1963) is an American businessman, electrical engineer, and the Co-Founder, President and CEO of Nvidia Corporation. As of February 2024, Huang's net worth was estimated to be $60 billion by Forbes, making him the 24th richest person in the world.

The incredible thing about Jensen—even rocking his patented leather jacket—is how unassuming he is. The man drives a Toyota Supra. Before last months earning call the Goldman Sachs trading desk called Nvidia “the most important stock in the entire world”. After the earnings call Nvidia had added the market cap of Ford, Ferrari and GM—combined.

Today’s lesson will be one of ambition, tenacity and survival. My aim is to not only demystify—as much for myself as anyone—Jensen and Nvidia, but to show why Jensen should be spoken of alongside both the greats who built our past, and also the small handful in line to build our future. He is the man who is figuratively, and quite literally, holding all the chips.

Early years

Jensen, or Jen-Hsun, was born in Feb, 1963, in Tainan, Taiwan. He and his family moved to Thailand when he was five, before he and his brother were shipped off to live with an uncle in Tacoma, Washington. Jensen was growing up on US soil from the ripe of age nine.

“I tried to impress her—not with my looks, of course, but with my strong capability to complete homework,” Jensen would say of the courting process between he and Lori.

Jensen’s next move would be what all young, smart, technology obsessed graduates do—a master's degree in electrical engineering at Stanford University, followed by a move into a role at semiconductors company, LSI Logic.

Pre-Nvidia … pre-iN-video-games

While at LSI Logic, Jensen learned how to master the intricacies of custom semiconductor and ASIC technology, honing his skills in large-scale integration and pushing the boundaries of chip design. He also gained experience collaborating with cross-functional teams and understanding the evolving needs of the electronics industry.

His next move was to Advanced Micro Devices, Inc., more commonly known as AMD, where he delved deeper into microprocessor design.

AMD was a semiconductor company based in Santa Clara, California, born of Jerry Sanders, and seven of his colleagues from Silicon Valley famed, Fairchild Semiconductor. Recently, Jensen told the Acquired Podcast that he loved him time at AMD, and in an alternate reality, he could see himself still there today.

1993: Jurassic Park, Chicago Bulls & the birth of Nvidia

The year of 1993 was amazing culturally. If you—like me—versticulated (*original word) at the intersection of movies, sports and video games then you were in heaven. 1993 was a mecca for jocks and nerds alike.

The two Sun engineers that Jensun met with on that fateful evening in 1992 were Chris Malachowsky and Curtis Priem. The three co-founders envisioned the next wave of computing would be in the realm of accelerated computing, specifically in graphics-based processing. Graphics for computers—and video games. Think Doom & Wolfenstein but in 3D.

With around $40k in the bank the team, set our to build the next generation of computer gaming. Their first product, the NV1 was successful, but only to a point. It lacked the industry support—mainly from game developers— that it needed to be successful due mainly to the fact that it was optimised for processing quadratic primitives, rather than triangle primitives, which was industry standard.

If you are lacking a PHD in computer science, what I am saying in layman’s terms is; they build something new … and people didn’t like it.

This little chip, and the impact it had on Nvidia, might be the one product that meant today we haver the ability to argue with ChatGPT about whether pineapple belongs on pizza.

The RIVA 128, or the Real-time Interactive Video and Animation accelerator, was so popular that it sold a million chips in the first four months of it hitting the shelves.

When Nvidia release the RIVA, in August 1997 it was down to it’s final one month of cashflow to fund payroll. It quite literally saved Nvidia from being nothing more than a failed also-ran.

“If you lose my money, I will kill you”

One of the funniest stories of the early stages of Nvidia was related to the late Sequoia Capital founder, Don Valentine. Don was a large, imposing figure in Silicon Valley circles and was famous for his no-nonsense approach to investing.

The story of Jensen and Don, is that when raising it’s first round of venture capital the two legendary figures met for Jensen to pitch the opportunity to invest in Nvidia. Jensen, although usually quite confident, smart and charismatic—totally bombed the pitch.

And Don told him about it. He told him it was terrible. Luckily for Jensen though, he was introduced by his former employer, LSI Logic's founder Wilf Corrigan. Wilf had spoken incredibly glowingly of Nvidia, and Jensen in particular, so much so that Don knew he was going to invest in him anyway.

As they were finishing their meeting Don reached out to shake Jensen’s hand and tell him he’d be investing. With the firm grip of a venture capital titan, he looked Jensen square in the eyes and finished with; “If you lose my money, I will kill you”.

Jensen today, is alive and well and Sequoia’s original investment in Nvidia is regarded as one of the great investments in technology history.

The middle years & the rise of the GPU

Riding on the back of the RIVA 128’s success, Nvidia went public on January 22nd, 1999. Later that year, Nvidia would launch a new product; the GeoForce 256. This product is the first product marketed as a GPU.

What was so great about the GPU? Well the GPU was the world's first Graphics Processing Unit (GPU). The GeForce 256 was a big advancement in graphics technology. It integrated both 2D and 3D graphics rendering, with hardware acceleration in a single chip. This had not been done before.

For a lot of us, that have not lived through all the previous technology waves, we may look at Nvidia, as somewhat of a new name on the global stage. But look at this historic revenue run-up of Nvidia’s starting a quarter-century ago.

Nvidia, by this stage, had begun to make a name for themselves in the niche of gaming. And game developers knew that their rabid fanbase would never be happy with the current graphics stack. It could never be more life life. Because of their reputation, Nvidia began working with Microsoft to develop the graphics for the Xbox.

The continued rise of gaming super-charged the company’s growth through the early 2000’s—but that wasn’t all Nvidia were good for at the time—CAD design systems, automotive infotainment, tablets and smartphones, video processing—all these use cases were built on top of Nvidia’s chips.

But just like all good drama stories, things were about to get a lot darker and stormier for Nvidia. Although, Nvidia was the little engine that could of the tech sector the years 2005 and 2006 saw negative, followed by stagnant revenue growth. The market was becoming increasingly saturated, and they were facing stiff competition, which was pushing down margins and slowing their growth overall.

A brief bright spot was in 2007, when Forbes named Nvidia its Company of the Year, highlighting its accomplishments in the early part of the decade. That same year, Nvidia launched CUDA (Compute Unified Device Architecture), a parallel computing platform that marked a significant shift in the capabilities of GPUs. However, unbeknownst to Jensen and his team, the storm clouds of the Global Financial Crisis were gathering, with the trigger being just a few subprime mortgages away.

2008 and CUDA

In the first report of 2008, amid the turmoil of the GFC, Nvidia took a write down of $200 million dollars of it’s first-quarter revenue. ‘Abnormal failure rates’ apparently the culprit. During this period, from a 2007 high to a 2008 low, Nvidia lost roughly 80% of the value of it’s stock. For most, if not all companies, an 80% drop in your stock price is an existential shock to your system.

Again, Nvidia would need to rebuild, and rebuild themselves. They would need to bet the house—again. It’s not something that particularly scared Jensen and the team. During the 90’s the had come up with an unofficial motto that; "Our company is thirty days from going out of business". That being said, people were invested. Their careers, their emotions, their finances. So they set out, to continue to build upon their totally new computing platform: CUDA.

In machine learning environments, CUDA's was a game-changer as well. ML is all about crunching massive datasets—a heavy lift—computationally speaking. But with CUDA devs can train and run models way quicker than with just CPUs. This boost has led to some mind-blowing breakthroughs, from recognising images and speech to processing natural language and beyond.

CUDA was thought of as the best-in-class software for many of it’s use cases. And Jensen knew it. This was one of the times in which he would pull one of the great levers in business. When you know you have a significant technological advantage—raise your prices.

This is a muscle Nvidia and Jensen have been building over time, and flashing forward to today for a moment, Nvidia’s current gross margins are at 70%, up from 66% only two years ago. Bananas to think about.

30-year overnight success

There are a lot of things that happened in the years in and around the crash of 2008. Competitor AMD would approach Jensen for a potential merger. Nvidia stock would plummet 50%+ (again) after a bad stock call. And deep learning would be supercharged with AlexNet, the brainchild of Alex Krizhevsky, Geoffrey Hinton and Ilya Sutskever of OpenAI.

Smart Silicon Valley insiders saw it coming though. Marc Andreessen in 2016 noticed that ‘everyone was building on Nvidia’ and was quoted as saying; “if we were a hedge fund we’d put 100% of money into Nvidia”. And NVIDIA's slow but strategic shift towards AI, deep learning, and data center markets puts it front row centre to capture the value of the next wave of computing.

But even with the strong tailwinds towards the end of the 2010’s, now one in their right mind could predict what would come to pass in the latter part of 2022.

The great battle for chips

Every so often their is a hype cycle in technology that blows the doors off the industry. You can go as far back as The Space Race, through the rise of the personal computer, the internet and the Dot-Com boom. More recently we have seem social media, mobile, blockchain & (lol) web3.

But there may be no bigger hype cycle—and no more meaningful one—than the cycle we are going through today. The hype cycle that is artificial intelligence.

On November 30, 2022, Sam Altman, and the team at OpenAI dropped a product release so popular and so revolutionary that it surprised even their staunchest believers—ChatGPT. The trajectory of Nvidia, and the world for that matter would be changed forever.

It was as if the man with the starters gun pulled the trigger and the race was on. Not only a race between big tech—Google, Amazon, Meta, and Microsoft—but a race against regulators, a race China, and if recent reports are too be believed a $7 trillion dollar race against the face of the AI revolution, Sam Altman himself.

One thing we do know, is that the market believes the story of Nvidia, with the stock commanding a 96x price to earnings ratio, coming off the heels of their latest earnings. “I’ve never seen anything like this”; Scott Galloway would say of their rise.

Before we take a final look into the future for Nvidia, let’s look at a few things that make them unique.

Rule #1: Bet the farm, just don’t die doing it

One thing that Nvidia has done really, really well since their founding in 1993 is what one could only describe as—not dying. Somehow, through Jensen’s phenomenal leadership they have managed to make bold, courageous bets, and have them miraculously pay off.

And although it may look lucky—and let’s be clear, there is a healthy amount of luck involved in any successful business tale—their success is also by design. Jensen is famous for talking about making sure Nvidia is in fact in the right place at the right time. Or close enough to be able to have impacts. He calls this the “zero-billion-dollar market”.

In order to be positioning yourself in this way, you need to be spending time on things that may not look important at the time. And you may need to push all your cards in. Take NVIDIA’s vision in ~2010 to create “neural network processors, safety architectures that run AI algorithms”.

This was at a time would have been close to a ‘zero billion dollar market’

On management & leadership

Jensen, has a unique perspective when it comes to management. And prefers to think from first principles when it comes to organisational design. Nvidia has a famously flat structure, unlike most companies you would see today.

Now, when I say a flat structure, I mean very flat. Jensen personally manages ~40 direct reports today. His reports submit a weekly list of the five most important things they are working on, of which Jensen reviews every morning. He elaborates further:

When talking about the typical org structure you see today—think C-level, VP, Head of this, Director of that—he was particularly critical, preferring a ‘horses-for-courses’ approach to organisational design.

I could go on. There are no departments, no business units, no status reports. They do not 1-year plan, 3-year plan, god forbid, 5-year plan. Jensen has a very particular way of running his org and hey—maybe we are the weird ones.

Playbook

Future

Today, so much of what we touch is powered by Nvidia. Your favourite social feeds, the algorithm you use while scrolling YouTube, even your car if you happen to drive a Tesla. And while the rise of Nvidia has been a 30-year success story, in Jensen’s eyes, they are just getting started.

In the history of technology, there has been one guiding principle regarding the pace of computing growth: Moore’s Law. But like all true rebels, Jensen sees this law as nothing more than an antiquated relic of less ambitious times.

It's the year 2024, folks. AI is here. The future has arrived. And if Jensen Huang has anything to do with it, we're only just getting started.

Fun factoids

Extra reading

And that’s it! You can also hope you enjoyed it folks. You can go out and build your own AI startup on Nvidia’s chips here, haha.