Shulou(Shulou.com)11/24 Report--

There is no doubt that the most popular semiconductor company this year is Nvidia, which has a market capitalization of more than 1 trillion yuan.

Unexpectedly, Lao Huang has a special, or even crazy, management style:

No plan, no report, no clear hierarchy.

The exposure says:

He directly manages 40 subordinates and believes that flattening is power.

It will be too late to wait for someone to report, so we have to do "random sampling" to the system of the company.

The information is completely transparent, from VP to junior employees can attend any meeting.

Plans are not made periodically, there is no five-year plan, and there is no one-year plan.

It sounds a little unexpected.

And Nvidia Deep Learning Research VP (@ Bryan Catanzaro) has directly appeared to prove:

It's all true.

Netizens commented that his management method was "like a GPU" and won the highest praise.

Lao Huang's crazy management rule has as many as 40 direct subordinates and does not communicate at 1:1.

Lao Huang's first management philosophy is to flatten and avoid hierarchical communication.

He has as many as 40 direct subordinates, but avoids one-on-one meetings.

Instead, group discussions have the advantage of ensuring that everyone (mastery of information) is on the same line from the top.

And he attaches great importance to learning and keeping abreast of the latest trends in the industry, and one way to do this is to surround himself with knowledgeable people who are willing to share.

Another aspect of flattening is that Nvidia's meetings are not restricted by position or level.

Lao Huang revealed that everyone, from vice presidents to junior employees, has access to all the company's information and has the right to attend any meeting.

His idea is:

If we have a strategic direction, why is it only known to a few people?

Tell all the employees, and we can put our heads together (original words: swimming in the soup of strategizing and how to formulate the path to the future).

We just send it to everyone as soon as we have an idea, and then wait for everyone to give us feedback, and then we improve it based on that feedback.

Only share the strategic direction, not make a clear plan

To share the strategic direction with every employee in the company is to use the wisdom and expertise of the entire company to develop a better and more thoughtful strategy.

This also led Lao Huang to admit that he had opened the door to getting feedback from different angles.

However, you need to pay attention to the above language: strategic direction.

Yes, it's just the direction. Lao Huang never makes a clear plan for the company, whether it's long-term or short-term.

The reason is that the plan is rigid and easy to bring restrictions.

The world is a living, breathing thing.

So Nvidia's approach is to constantly re-evaluate existing strategies in the light of changing business and market conditions.

Given the rapid development of the AI field today, this is indeed a very wise move.

Besides dislike making five-year plans and one-year plans, Lao Huang also doesn't like employees to write reports.

Because he believes that such reports are often divorced from the basic facts by the time they reach him.

But he still needs to keep abreast of the company, so he encourages all employees to email him with the subject of the "five most important things" on their minds.

Every morning, Lao Huang reads about 100 such emails.

Jump out of the rules and dare to give up existing business

All of the above are actually related to how to better collect information and obtain strategies.

So, how did Lao Huang formulate his strategy?

The answer is Think Outside the Box.

A few years ago, Nvidia pulled out of the already commercialized smartphone and tablet SoC market.

Now it seems that this seemingly backward move has propelled the company forward and made Nvidia stand out from its competitors.

What is reflected behind this is Lao Huang's strategic philosophy:

Instead of following the established practice, everything builds a solution from the facts on the premise of disassembling the problem as a basic fact.

Lao Huang said that this approach can encourage the birth of innovative thinking and often lead to breakthrough solutions that traditional methods may ignore.

It is worth mentioning that Lao Huang also likes to trust his intuition-this is another way for him to "Think Outside the Box".

Three core goals

Finally, Lao Huang summed up Nvidia's organizational structure to achieve the following three core goals:

Attract the best talent in the industry, set up the most capable and efficient team to operate, and in the process, ensure the rapid and efficient flow of information throughout the company.

On the whole, these management methods are indeed a little unique and crazy.

But it is not hard to see that its core feature is to avoid the bureaucracy and rigidity common to large companies by retaining the flexibility of start-ups.

One problem is:

Nvidia is still growing rapidly. As the company grows in size, can this streamlined management style still live in hold, and can its successor continue?

Lao Huang: after it was exposed that I could fight the crazy management style of Lao Huang in 20-30 years old, netizens also started a heated discussion.

Some netizens from Nvidia confirmed:

After only a few months of working in the company, I began to discuss the project at hand with Lao Huang face to face.

Working under his "microscope" is stressful, but it greatly reduces the nonsense of showing a beautiful PPT and saying "everything's fine" to the leader.

At the same time he recalled:

I've worked at a start-up that is only the size of Nvidia 1max 100, and there is much less connection between CEO and an ordinary engineer.

So this is a very good change.

However, some employees also revealed:

It's just Lao Huang who doesn't report at 1:1. Org3 employees like us still have regular one-to-one communication with my Org2 leaders.

So is that why Nvidia is successful?

@ Dan Hockenmaie believes that:

It's hard to know. But if Nvidia hadn't developed at this speed and flexibility, it might have died several times early on.

It is worth mentioning that some netizens are also worried:

After Lao Huang retires, can Nvidia survive like this?

In response, @ Dan Hockenmaie revealed:

It's too early to worry. Lao Huang's plan is to fight for another 20-30 years.

Reference link:

[1] https://x.com/ctnzr/status/1701736726908240311

[2] https://www.tomshardware.com/news/nvidia-ceo-shares-management-style-always-learn-make-no-plans

[3] https://news.ycombinator.com/item?id=37486882

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.