看不懂的sol哥

2026机会多多

3Following

2.5Kfollowers

Feed

Livestream

Feed

Jensen Huang, how did he learn? (Long read, proceed with caution) Today, NVIDIA has long dominated as the world's most valuable company. Many people think its founder Jensen Huang saw the future from the very beginning. But Jensen Huang once said that when NVIDIA first started, its technical path was wrong. What was worse, the team realized the direction was wrong, but no one knew the right answer. At that time, Jensen Huang had only a few hundred dollars in his pocket. He went to an electronics store, bought three textbooks, and handed them to the engineers so everyone could start learning from scratch. This sounds a bit absurd, but NVIDIA’s ability to get where it is today relied on this capacity to relearn. Today, let's talk about how Jensen Huang turned himself and the company into a "learning machine." 1. How did NVIDIA survive after starting off on the wrong foot? In 1993, Jensen Huang started a company with two partners. Their idea was simple: create a dedicated acceleration chip to allow ordinary computers to smoothly run complex 3D graphics that previously only large workstations could handle. The idea was attractive, but reality was harsh. They bet on the wrong technical path; the chips they made were expensive, development was costly, and the technology was not mainstream. At first, they stubbornly believed that as long as the technology was strong, the market would eventually understand. But the market gave them no mercy; dozens of competitors were fiercely fighting for territory, and the company’s finances were nearly depleted. What was worse, NVIDIA was still carrying a chip order from Japan’s Sega. Jensen Huang knew clearly they couldn’t deliver. He bravely flew to Japan, spread his hands, and told Sega’s executives: "Sorry, we messed up, chose the wrong path, please find another supplier quickly." Then he said: "Although the work isn’t done, can we still keep the contract payment? Without this money, NVIDIA will go bankrupt tomorrow." Admitting failure was humiliating enough, but asking for life-saving money took great courage. Jensen Huang didn’t play tricks; he sincerely explained the company’s desperate situation and begged for one last chance to find a way out. Sega was moved by this honesty and gave them a break. NVIDIA used the lifeline to start over and created the life-saving RIVA 128 graphics card, finally stabilizing on the edge of the cliff. True learning often doesn’t happen after smooth success but at the moment reality slaps you hard. 2. What is true strategic resolve? Distinguishing "goals" from "means" If NVIDIA had just survived with the RIVA 128 and quietly sold graphics cards, today it would at best be a mediocre hardware company. Its transformation into a giant lies in one hard truth Jensen Huang firmly believed: the tactics in hand can always change, but the core problem to solve must never be abandoned. 3D graphics was just a stepping stone. Later, NVIDIA pushed into scientific computing, image processing, and today’s deep learning. The most typical example was the development of the CUDA platform. NVIDIA invested astronomical amounts of real money, but the market was cold and developers didn’t buy in. Shareholders were furious, thinking it was a money-burning waste. But Jensen Huang saw beyond short-term losses to a simple underlying logic: this chip, originally only for graphics, would one day help humans solve more complex general computing problems. He endured many years, not because he predicted exactly when AI would explode, but because he was convinced of the long-term value of "accelerated computing" for humanity’s future. When the deep learning era arrived, NVIDIA’s years of hard work instantly found its use. This is the essence of "strategic resolve" versus "blind stubbornness": Stubborn people cling to past solutions out of pride and fear of admitting mistakes; Those with resolve cling to the core problem to solve, while tools, methods, and markets can be changed anytime. This is also why NVIDIA dared to decisively cut its mobile chip business. The mobile market seemed a goldmine, but its extreme demands for low power and low cost clashed with NVIDIA’s core gene of high computing power. Painfully giving up a huge market hurts revenue and morale. But for Jensen Huang, there are hundreds of profitable markets; NVIDIA’s mission is only one—to help machines solve more complex problems. Since the mobile market deviated from the main line, they would not linger. 3. A CEO must never "sit in the VIP seat and watch the show" In many big companies, bosses usually only look at summary PPTs, but Jensen Huang still likes to get his feet in the mud. He reads cutting-edge hardcore papers himself, talks directly with scientists about fundamental principles, questions why a technical detail matters, and repeatedly confirms with frontline engineers exactly where customers are stuck. Even with the company’s market cap at the top, he still needs to understand how their products are built line by line of code, chip by chip. Some say this is micromanagement and lack of delegation. Jensen Huang says: A CEO must serve the company and provide employees with new understanding so they can act with confidence. He uses surfing as a metaphor: an outsider sees only chaotic waves; a true surfer sees the rhythm and direction, timing the perfect ride. Facing rapidly changing tech waves, outsiders see a mess; those who understand fundamentals see a clear framework. Founders must stand on the front line to feel the water temperature, let professionals do their jobs, but keep eyes on the wave’s direction to ensure the ship doesn’t fall behind. He also likens NVIDIA to an F1 race car he personally drives. The company’s structure can’t copy business school textbooks; it must be born for real goals. A driver doesn’t tighten every screw but must hold the wheel and feel every vibration of the tires gripping the road. NVIDIA’s extremely flat structure follows this logic. Jensen Huang believes the company tackles extremely complex engineering problems; information can’t be filtered through layers. If a problem has to circle through seven or eight reporting levels, the race car would have crashed long ago. Founders must never sit safely in the audience; they must be at the first scene of the problem, understand what’s really happening, then grip the wheel and lead the team forward. 4. How did Jensen Huang turn the company into a "learning machine"? Talking about Jensen Huang’s management, you often see buzzwords online: flat organization, "five key items" emails, whiteboard meetings, captain system, open feedback. Separately, none are extraordinary. But together, they focus on one thing: Letting real voices surface early, assigning responsibility to specific people, and turning mistakes into team muscle memory immediately. First, extreme flatness. Over 60 executives report directly to Jensen Huang, far more than other companies. Outsiders think he’s showing superhuman energy, but he cares about whether information is lost in the middle. In big companies, many layers turn bad news into polite words, fatal flaws into pretty KPIs, and department conflicts into "overall controllable." Shortening the information chain aligns intelligence instantly in everyone’s mind, letting managers grasp frontline changes anytime. But this approach requires high thresholds: subordinates must make independent decisions, leaders must withstand information bombardment, and the company must have a very clear big goal as a compass. Without these, flatness only leads to endless bad meetings and chaotic direction. Next, "mission is the true boss." In traditional big companies with territorial divisions, employees protect their department’s interests first, caring little about the company. NVIDIA treats all employees as a shared talent pool; when a big battle comes, people move to that project, and department affiliation is secondary. This breaks deadlocks of departmental disputes. When cross-department projects fail, meetings focus only on customer pain points, data, and facts. Loud voices or high ranks don’t decide; power follows the task. After the battle, resources and people are reshuffled to tackle the next hard problem. "Five key items" emails. This cures the hypocrisy of daily reports. Many companies’ weekly or monthly reports have become workplace performances: carefully worded, hiding bad news, beautifully formatted, forwarded layer by layer until the real crisis disappears. Jensen Huang forces everyone to report the hottest urgent issues, clearly stating what customers are complaining about, where projects likely fail, and what competitors are up to. He focuses on these early "weak signals," which are a thousand times more useful than polished reports. "Whiteboard meetings" force you to reveal your thinking. PPTs are for decorating already-made conclusions; whiteboards require you to pick up a marker and explain from zero: what is the problem? Are your assumptions reliable? Why choose this path? What’s still unclear? Standing at the whiteboard, no one can bluff with a script. Everyone ruthlessly dissects logic, challenges flaws face to face. The meeting’s meaning shifts from "showing how well I did" to "solving the tough problem together." At NVIDIA, even the most complex cross-department projects must have a single "captain." The captain may not be the highest rank but is fully responsible for the final result, holding decision and resource power. If things fail, no excuses like "department coordination issues"—everyone watches how the captain pulls the plane up. Jensen Huang highly values "face-to-face, timely feedback." Problems are exposed immediately so the team can turn around and fix them; public reviews let the whole company understand how the crash happened, preventing repeats. But this double-edged sword has limits. Poorly handled public criticism can become humiliation and workplace fear. This strict approach works at NVIDIA due to high talent density, unified passionate mission, and deep trust built from years of victories. These cycles naturally make NVIDIA a continuously evolving learning machine. 5. Learning while taking hits Whenever Jensen Huang talks about personal growth, he mentions pain, low expectations, and resilience. If you see this as just "no pain, no gain" motivational talk, you underestimate him. He means: when you’re in the dark, things are a mess, do you still have the spirit to keep learning and pushing forward? Pain itself isn’t to be glorified; the key is whether it leads to clearer next moves. Once you realize you’re off track, you must quickly admit it and change course. Clinging to past glory in a changed environment, insisting "I can’t be wrong," is suicidal. Admitting mistakes is recognizing reality: your past experience was just luck. When facing a completely new field, don’t wait—learn enough to "get your hands dirty" first. Jensen Huang didn’t sit in his office waiting for top chip experts to parachute in; he grabbed a few hundred dollars, bought textbooks, and led his team to study all night, building the basic framework before fighting real battles. We do the same: focus on the most critical bottleneck, learn enough to get the wheel rolling, and fill other gaps through hands-on struggle. Today’s AI makes gathering knowledge easy. What scarce skill will be valuable in the future? Can you find the learning target? Can you untangle a mess into a clear framework? Can you embed learned insights deeply into reality to solve real problems? Jensen Huang often says: "How hard can it be?" Does he really think it’s easy? Of course not; he knows the path ahead is a mountain of blades and fire. But if someone hasn’t taken the first step and has already imagined all the ways to fail in the next ten years, they’ll just lie down and give up. The most solid approach is to focus on the ground beneath your feet and fight hard. Today, grit your teeth to understand an unfamiliar formula; tomorrow, visit clients to clarify pain points; the day after, push the team to release a new code version. Action is the only weapon to smash inner demons and make the unknown concrete. In the raging AI wave, the strongest confidence is to break arrogance, roll up your sleeves, and believe you can "learn while taking hits all the way through."