Jensen Huang
NVIDIA
Core Principles
culture
Criticize publicly so the entire organization can learn from a single person's mistake. Optimize for company learning, not for protecting individuals' feelings or comfort.
Jensen will call out engineers at all-company meetings and doesn't use private feedback. He once asked if a chip was 'the piece of shit you intended to build' in front of the team, and invited a Best Buy executive to publicly detail product failures. He frames this as refining company character, not as personal attacks.
“Feedback is learning. For what reason are you the only person who should learn from this? We should all learn from that opportunity. We're not optimizing for not embarrassing somebody. We're optimizing for the company learning from our mistakes.”
focus
Ruthlessly prioritize work by identifying your highest priority activity and completing it before the workday begins. Focus all energy on what is truly essential and ignore lower-priority activities entirely.
Jensen starts each day by completing his most important work before helping others. He has a very clear priority list and works from highest to lowest priority. Once his top priority is complete, his day is already a success, freeing him to support others.
“I have a very clear priority list and I start from the highest priority work first. Before I even get to work, my day is already a success.”
hiring
Attract top talent through exceptional compensation, including equity grants that can be awarded throughout the year rather than only during annual reviews. Recognize extraordinary contributions immediately.
Jensen reviews stock allocation reports personally and awards special one-off RSU grants to top contributors outside of normal review cycles. When a grant is approved, the employee receives an email recognizing their extraordinary contributions. This approach ensures high performers feel appreciated and compensated in the moment.
innovation
Believe that constant reinvention is mandatory, not optional. Successful ideas must eventually be erased and replaced with new ones. A company is never more vulnerable than at the height of its success.
Jensen views the whiteboard as representing both possibility and ephemorality, the requirement that brilliant ideas must be erased. He believes NVIDIA's survival depends on constant reinvention, not on defending past successes.
leadership
Your reputation will precede you. Protect your long-term reputation above all else because it determines whether you get the best opportunities and whether people believe in you.
Don Valentine invested in NVIDIA based on Jensen's reputation, not on the strength of Jensen's pitch, which he called terrible. Jensen realized that his reputation alone made people trust him enough to invest. This lesson stayed with him throughout his career as an essential asset.
Seek conflict and confrontation, not agreement. Constructive conflict refines ideas and yields better results than consensus building.
At LSI Logic, Jensen observed that explosive fights and arguments about what to build actually led to better decisions. He understood the concept of honing the sword, where rough surfaces rubbing together create a sharper blade. He adopted this as a core NVIDIA philosophy.
Jensen is a teacher first. The best leaders spend most of their time teaching the organization, creating alter egos who understand the mission as well as the founder.
Employees at NVIDIA call him Professor Jensen because he excels at explaining complex concepts simply. He uses whiteboards constantly to teach and diagram ideas. His core philosophy is to create alter egos throughout the organization who embody his values and thinking.
Design a flat organizational structure to fight complacency and slow decision making. Flatten hierarchy so information travels quickly and employees are empowered to act.
Jensen realized that NVIDIA became bureaucratic and political as it grew. He deliberately created a flat structure with no middle management layers. This allows information to flow freely and prevents the organizational rot he observed at other companies.
“By having a lot of direct reports, not having one-on-ones, we made the company flat. Information travels quickly. Employees are empowered.”
Build your company as an extension of yourself. Design the organization structure and culture to match your natural strengths and working style so you can sustain it for decades.
Jensen built NVIDIA to be managed directly by him with 60 direct reports and no one-on-ones. This flat structure reflects his preference for speed and information flow. He refused to change this even when board members suggested hiring a COO. The company is Jensen with 29,000 lives.
“The company's organization is like a race car. It has to be a machine that the CEO knows how to drive.”
marketing
Simple, clear messaging is essential for seduction. Make your positioning and message so simple that anyone can understand what you do and who it is for.
Jensen learned that seduction requires a simple message. His early pitch for NVIDIA was too complicated and unclear. This lesson drove him to obsess over clarity in how NVIDIA positioned itself and communicated with customers and partners.
“Seduction requires a simple message.”
Educate the market when customers don't yet understand how to use your technology. Conduct seminars, write textbooks, and teach at universities to establish your technology as an industry standard.
NVIDIA offered CUDA-capable machines to schools that committed to teaching the subject. The chief scientist gave over 100 talks in one year and wrote a textbook that sold tens of thousands of copies in multiple languages. This education effort attracted talent and drove adoption of CUDA as a standard.
mindset
Ruthlessly criticize yourself in public to force continuous learning. Don't shield yourself from harsh feedback or self-judgment, as this is what drives improvement.
Jensen was embarrassed by how little he understood about positioning the NV1. Rather than hiding this, he openly acknowledged the failure and used it to learn what he needed to know. This public self-criticism became a model for how he managed NVIDIA.
Be unapologetically extreme in your work ethic and standards. No one will outwork you, and this commitment creates a competitive advantage others cannot match.
Jensen works every single day without vacation. He will grill employees at the urinal about what they are working on. He recruits by telling people he will put competitors out of business. He tells employees that long hours are a prerequisite for excellence, like Olympic training.
“There may be people smarter than me, but no one is ever going to work harder than me.”
Believe that your company's worst enemy is not the competition but your own complacency. Build organizational systems and culture specifically designed to prevent the arrogance and inertia that grip successful companies.
Jensen repeats 'complacency kills' throughout the company. He tells employees monthly that they are 30 days from going out of business. He resists positive narratives about early success and maintains this paranoid mindset even as NVIDIA becomes the world's most valuable company.
“Complacency kills. NVIDIA's worst enemy is not the competition, but itself.”
Develop a habit of relentless self-criticism. The ability to endure pain and suffering, combined with the capacity to see opportunities in setbacks, are greater superpowers than raw intelligence.
Jensen looks in the mirror every morning and says to himself 'you suck.' An executive was struck that someone so manifestly successful could hold such a view. Jensen frames suffering as essential to building character and resilience, not as something to avoid or minimize.
“I look in the mirror every morning and say, you suck. My ability to endure pain and suffering. My ability to work on something for a very, very long period of time. My ability to handle setbacks and see the opportunity just around the corner, I consider to be my superpowers.”
Be unapologetically extreme in your commitment to work and winning. Work every hour, insist that no one will outwork you, and communicate directly that second place is the first loser.
Jensen works every day and thinks about work when not working. He tolerates no complaints about long hours, comparing it to Olympic athletes training. He told a potential hire that he would put Silicon Graphics out of business. He operates without apology from this extreme position.
“There may be people that are smarter than me, but no one is ever going to work harder than me. Working is relaxing for me.”
operations
Use blunt, concise, direct communication in all settings. Keep emails short and communications crisp, like a haiku, because compression of thoughts shows discipline.
Jensen's emails are famously short and to the point. He believes long communications are the enemies of speed. He reinforces this with the LTUA acronym: listen to the question, understand the question, answer the question.
Restructure your operations to match market rhythm, not engineering convention. Split teams so you can release products in parallel cycles, not sequential ones.
Jensen realized competitors released chips every 18 months while PC makers refreshed every 6 months. He split NVIDIA's design teams into three groups working in parallel, allowing them to release a new chip every 6 months. Three teams, two seasons made NVIDIA impossible to compete against.
“Three teams, two seasons. Even if a competitor offered a slightly better product, PC makers would have no motivation to switch away from NVIDIA knowing that a faster part would arrive within six months.”
Invest heavily to shrink cycle time, not just to add features. Buying a million-dollar emulator to shorten testing by 3 months can save your company.
When NVIDIA was running out of money, Jensen bought an EICOS emulator for a million dollars to cut the testing cycle from 9 months to 6 months. This expensive investment paid for itself by getting Riva 128 to market faster, which saved the company.
Use the whiteboard as your primary communication tool in meetings. It forces transparent thinking in real time and prevents people from hiding incomplete ideas.
Jensen uses whiteboards constantly in meetings, even when others are speaking. He has a favorite brand of marker only sold in Taiwan and travels with whiteboards. The whiteboard forces rigor, transparency, and requires people to start from scratch each time.
“At the whiteboard, there is no place to hide.”
Implement a top five things email system to intercept weak signals before they become problems. Read direct reports from all employees to stay connected to what is happening at the edge.
Every NVIDIA employee at all levels must send a weekly email listing the top five things they are working on, observing in the market, or customer pain points. Jensen reads over 100 of these daily. He caught the weak signal of machine learning years before it became the AI boom.
“Strategy isn't what I say, it's what they do. I want information from the edge.”
Frameworks
LTUA (Listen, Understand, Answer)
When someone rambles or loses focus, say LTUA. It means listen to the question, understand the question, answer the question. It is shorthand for disciplined thinking and concise communication. Use it to interrupt fuzzy thinking in meetings.
Use case: Meetings where people are not being crisp or direct. Use as a gentle redirect to force clarity.
Honing the Sword
Seek out conflict and confrontation in your work because rough surfaces rubbing together create sharper results. Explosive fights and arguments about strategy, product direction, and execution lead to better ideas than consensus building. Like sharpening a sword on a wet stone through repeated friction, teams polish each other and their ideas through constructive conflict.
Use case: Team meetings where strategy or product decisions need refinement. Apply when consensus feels too easy and you suspect ideas haven't been tested rigorously.
Speed of Light
Calculate the theoretical maximum speed at which a project could be completed by assuming no delays, queues, or downtime, constrained only by the laws of physics. This becomes your performance benchmark. Forbid teams from comparing themselves to past performance or competitors. The speed of light is the only comparison.
Use case: Setting aggressive timelines for projects. Use this to break teams out of incremental thinking and force them to identify all wasted time.
Top Five Things (T5T) Email System
Every employee at all levels sends a weekly email with five bullet points starting with action verbs, describing what they are working on, observing in the market, or noticing at the edge of the business. The leader reads all of them. Tags by department and subject line make them searchable. This flattens hierarchy and gives the leader unfiltered information from the edge.
Use case: Organizations at 50+ employees where the CEO starts losing visibility into what is happening. Implement this to catch weak signals before they become problems.
Three Teams, Two Seasons
Divide your design or product team into three groups working in parallel. One team designs the new architecture while two others design faster derivatives based on it. This allows you to release products on a faster cadence than competitors who work on one chip at a time.
Use case: Industries with rapid refresh cycles (hardware, software releases). Use when competitors have longer development cycles and you want to make staying ahead of you impossible for them.
Mission is the Boss with Pilot in Command
Organize the company by functions (sales, engineering, operations) not divisions or business units. When a new mission emerges, assemble all functions under a single pilot in command accountable for that mission. This person has a name and reports directly to the CEO. This structure allows rapid pivot when strategy changes.
Use case: Companies that need to shift from one market to another (like NVIDIA from gaming to AI). Use when you want functional flexibility without reorganizing the entire company.
Professor Jensen Teaching Method
Leaders communicate complex concepts on whiteboards in meetings, explaining strategy and vision so consistently that employees at all levels understand and can articulate the same ideas. Information flows directly from leadership rather than through management filters, creating organizational alignment around shared mental models.
Use case: When you need to ensure that strategic vision is understood and internalized across a large, growing organization.
Speed of Light Performance Benchmarking
Break all work into component tasks with target completion times that assume zero delays, queues, or downtime. Measure performance against the theoretical maximum speed possible (limited only by physics), not against past performance or competitors. This sets an objective external standard rather than a relative one.
Use case: When you want to prevent complacency and incremental thinking, and push teams to maximize efficiency.
Top Five Email System
All employees at every level send weekly emails with five bullet-point action items detailing what they are working on and market observations (customer pain points, competitor activity, technology developments). Leadership tags emails by topic for easy searching and reads a sample daily to get unfiltered edge information.
Use case: When you need to maintain visibility into the entire organization, intercept weak signals early, and prevent information from being filtered through management layers.
Pilot in Command Accountability
Every project and initiative must have a named leader (pilot in command) who reports directly to the top and owns the outcome. The mission is the boss. Responsibility cannot hide behind departments or committees. Each person's name is attached to their work.
Use case: When you need to ensure clear accountability, eliminate diffusion of responsibility, and maintain focus on mission rather than organizational hierarchy.
Stories
Kenneth Hurley was standing at a urinal when Jensen walked up next to him. Jensen asked what he was working on, and when Hurley panicked and thought he was about to be fired, he rapidly listed 20 things he was working on. Jensen replied 'Okay' satisfied with the answer. No place in company headquarters was safe from Jensen's drive-by grilling.
Lesson: Extreme leaders find opportunities to engage and probe everywhere, and employees cannot hide from accountability. Jensen's relentless engagement is pervasive throughout the organization.
An executive at NVIDIA describes a moment when the company had just blown the doors off quarterly results. Jensen stood up and said 'I look in the mirror every morning and say, you suck.' The executive was struck that someone so manifestly successful could think in such terms.
Lesson: The greatest founders combine extreme self-confidence with relentless self-criticism. Success does not satisfy them; they maintain an inner voice that says nothing is good enough.
In 2007-2008, Jensen invested so heavily in CUDA compatibility that NVIDIA's gross margins fell from 45% to 35%. The stock fell 80% during the financial crisis. Investors demanded a course correction. Jensen remained committed, saying 'I believe in CUDA. We are convinced that accelerated computing would solve problems that normal computers couldn't.'
Lesson: Conviction in long-term opportunities sometimes requires making short-term financial sacrifices. The greatest opportunities may require you to ignore investor pressure and short-term metrics.
Jensen was furious at poor planning and execution of a chip called the NV30. He called out engineers at an all-company meeting saying 'Is this the piece of shit that you intended to build?' He later invited a Best Buy executive to speak about the product's poor performance and customer complaints, publicly validating the criticism.
Lesson: Public criticism serves organizational learning. It also demonstrates that the leader's standard for quality is non-negotiable and that executives outside the company see the problems too.
Peter Young was introduced to Jensen at a party for new hires. Jensen already knew who he was and recalled Peter's career history, including his work at Sony PlayStation and 3DFX. Jensen had similar recall of biographical details for all 50 attendees.
Lesson: Leaders who excel are deeply involved in details and care enough to know about individual people. This attention creates connection and shows people they matter.
Jensen was recruiting a chief engineer from Silicon Graphics (SGI). He invited the engineer to lunch and told him directly: 'John, you should really think about coming to our company because ultimately, I'm going to put SGI out of business.' Jensen was unapologetically extreme about his intentions.
Lesson: Extreme confidence and directness can be attractive to top talent. Showing ambition openly and stating competitive intent directly communicates conviction.
Jensen was four years old when his father visited New York and realized America was the land of opportunity for his two sons. His parents sacrificed everything, selling their possessions to send Jensen to an American boarding school in Kentucky, thinking it was a prep school when it was actually a reform school for troubled youth. The school had strict routines and physical hardship, including getting beaten by older, larger students.
Lesson: Character is built through adversity and pain, not inherited intelligence. Jensen's early willingness to endure discomfort shaped his ability to tolerate the pain of building a company for 30 years.
While working at AMD, Jensen worked full-time as a microchip designer by day and attended Stanford for a master's degree at nights and weekends while raising a young family. It took him eight years to complete the degree while maintaining his full-time job and family responsibilities.
Lesson: Extreme long-term patience combined with impatience about certain things allows you to compound effort over decades. This capacity to endure sustained discomfort becomes a competitive advantage.
Jensen was working at LSI Logic when he was assigned the Sun Microsystems account and met two gifted programmers, Chris Malachowski and Curtis Preem. They became upset with Sun and wanted to start their own consulting business. They came to Jensen because they needed business help and loved working with him. Jensen convinced them to start their own company instead, becoming co-founders of NVIDIA.
Lesson: Opportunity handled well leads to more unexpected opportunity. Doing every job exceptionally well creates relationships and reputation that unlock future possibilities.
NVIDIA's first product, the NV1, was a complete failure. It was over-designed with features nobody wanted. The market simply wanted the fastest graphics performance at a decent price. Jensen felt embarrassed by how little he understood about positioning, but used this public failure as a learning opportunity.
Lesson: Don't hide from failure. Use it to learn what you need to know. This ruthless self-criticism forces growth and prevents repeating the same mistakes.
Notable Quotes
“I wish upon you ample doses of pain and suffering.”
When asked about his advice for achieving success. He believes greatness requires enduring adversity and pain.
“There may be people smarter than me, but no one is ever going to work harder than me.”
Explaining his competitive advantage and why he has built such a successful company.
“We learned it was better to do fewer things well than to do too many things. Nobody goes to the store to buy a Swiss Army knife. It is something that you get for Christmas.”
After the NV1 failure, realizing the market wanted simplicity and focus, not features.
“At the whiteboard, there is no place to hide.”
Explaining why he uses whiteboards constantly. They force transparency and rigor in thinking.
“By having a lot of direct reports, not having one-on-ones, we made the company flat. Information travels quickly. Employees are empowered.”
Describing the organizational structure he designed to prevent bureaucracy and slow decision-making.
“Feedback is learning. For what reason are you the only person who should learn from this? We should all learn from that opportunity.”
Explaining why he gives public criticism instead of private feedback.
“I don't like giving up on people. I'd rather torture them into greatness.”
Describing his leadership philosophy of maintaining relentless standards while believing in people.
“We are going to judge ourselves against the speed of light. I don't give a shit what other companies are doing.”
Explaining the speed of light framework. NVIDIA will only benchmark against physical maximums, not competitors.
“Strategy is not words, strategy is action. We don't do periodic planning. The world is a living, breathing thing. We plan continuously.”
Rejecting five-year plans and explaining NVIDIA's continuous strategic adjustment.
“Strategy isn't what I say, it's what they do. I want information from the edge.”
Explaining why he reads all T5T emails from employees. Real strategy emerges from actions at the frontlines, not from the CEO's words.
More Semiconductors Founders
Want Jensen's advice on your business?
Our AI has studied Jensen Huang's biography, principles, and decision-making frameworks. Ask any business question.
Start a conversation