Lu joined Google in 2006 after graduating with a master’s degree in computer science from Fudan University. Like most digital farmers, he had a dream of changing the world, but the path to that dream wasn’t clear when he first left school. In such a state of life, rather than lock the career track too early, it is better to join the forge ahead, innovative technology factory, there are five reasons:
First, there is so much to learn, with the shoulders of giants on the horizon and free Tech Talks to satisfy every knowledge-hungry soul.
Second, All In business growth, despise the innovation at the bottom of the commercial large companies, often expect employees to play the role of screws, each to perform their own duties, pulling the car. “Google’s default requirement is full stack engineer” is famous. “Light algorithm is not good, you have to be able to do engineering implementation, sometimes testing, monitoring, pressure measurement, on-line, all have to be done by yourself”, Lu Pengjun said.
Third, with the ship forward for a period of time, in the most advanced information ocean can more accurately establish the direction of life;
Fourth, get to know the industry’s top elite, not only three people must have my teacher, but also many friends multiple ways;
Fifth, on the road of rapid iteration, big factories that encourage innovation have great tolerance for mistakes.
One lesson lu remembers most: a $30 million loss for the company.
At the time, he was just a T5 engineer at Google, but because of his performance, he had a lot of interaction with the global VP of Google, and was expected to get a lot of access: limited access to HIP core code by many engineers, highly controlled user logs, and naked write access to Google’s database.
Lu Pengjun made a direct change to the database, which caused the failure of Google’s advertising system. The customized blacklist of customers’ advertising failed. As a result, a large amount of expenses were spent on the channels that customers did not want to put in, totaling up to 30 million DOLLARS.
Google’s inclusiveness has allowed Mr Lu to escape criticism. From this experience, Lu developed a unique philosophy and methodology for dealing with problems:
First, mistakes can be made, no one is perfect, but you need to dig deep to find the most fundamental trigger. After the loss was caused, Lu Pengjun had to face a deep interrogation in the Case study, “Why write directly? Why not go through the process?” “What are you thinking when the process is lazy?” “Why don’t you get someone to review it?”
Second, establish processes and mechanisms to get things right.
Third, review the process constantly to ensure that while solving problems effectively, it does not slow down rapid iteration and ensures that the process is efficient.
In the process of rapid iteration, exposing problems is not necessarily a bad thing for the whole system. Although he was severely tortured by the reexamination, this is the only so-called punishment lu Pengjun received.
Unfortunately, for all its deliciousness, There is one big problem: the lack of a business presence in China. With age, Lu pengjun, who is increasingly skilled, is more and more eager to devote himself to actual combat. In 2014, Lu joined Baidu along with many of his old friends, although he was repeatedly urged to stay by Google’s senior management. “Now I think I left a little late”.
Baidu period from 2014 to 2018 transformation management slightly earlier
With the recommendation of his senior friends, Lu pengjun became a director of Baidu only one year later, taking over the prestigious Phoenix Nest team in the Internet advertising industry.
Transformation management is the key step in the development of technical personnel. Before taking over Phoenix Nest, Lu had little management experience. Leading a team of 130 people was obviously a difficult challenge. However, Lu relied on his deep understanding of the business to successfully complete the transition. One year later, the company has more than 1,000 employees, leading phoenix Nest’s products, technology and operation.
When it comes to management transformation, Lu pengjun offers a surprising suggestion: technical people don’t have to rush to management transformation.
Kechuang: This is a rare suggestion. Could you share your specific thoughts?
Lu pengjun: Writing code allows you to think more thoroughly about a problem, which is very important in solving some core problems. Management in the Internet era is no longer purely about personnel management, but more about controlling direction and ensuring business success, which cannot be separated from strong judgment. After many people transfer to management, they will spend a lot of energy to coordinate resources, find ways to balance the interests of all parties, and constantly empower others. In this process, if not handled well, it can take people out of line, and over time, your judgment can be greatly affected. If you don’t know exactly what’s going on on the ground, you’re going to be very unreliable.
Science and technology founder: Could you describe the ideal working status of technical management position in your mind?
Lu Pengjun: I personally hope to be a “technical engineer with management ability”, who has both the right to make technical decisions and the ability to deal with core code; Even if you make the transition to a management position, you need to have a mechanism in place to make sure you clearly understand what’s happening on the front line. Just listening to the report is never enough. Some reports, the vocabulary is very advanced, sounds vaguely understood, the method is also very advanced, but can not be implemented. Therefore, after listening to the report, we must think positively, dig deeply for the problems, and constantly learn to enrich themselves.
Go all the way to promote business innovation
Lui felt a little regretful about “managing the transition too early”, but when asked whether “technical people should build business thinking as early as possible”, he gave a strong positive answer: “I think I left Google a little late, but I felt a little late to get close to the business”. May be out of door latecomers, LuPengJun for “how to vigorously promote business growth” showed the passion of 120 points, this also is many super time line growth is a common trait: clear objectives, not constrained by the method, that help customers will try pushing clients, rather than simply passive acquisition needs, meet the demand. Few people know that OCPC, which changed Baidu’s advertising model, was born of Lu pengjun’s desire for the honor of “Baidu’s top award”. In 2017, Lu Pengjun felt it was uncertain to win baidu’s top prize, so he held a brainstorming meeting to discuss countermeasures. One of the T10 engineers in the team proposed a bold hypothesis, “Payment by conversion rate can create value more directly than payment by click, so it should also create more considerable financial returns”. Lu Pengjun eyes a bright, immediately clapping, to promote, directly on the line. “There was a lot of opposition within the team. Some people even said, ‘It’s evil to change the price without talking to the customer’. One of the T9 engineers even quit because of this. But WHAT I see is that the general direction of conversion billing is definitely more effective. Maybe the current implementation method is not the right one, but if you wait for everything to be laid out properly, complete communication, complete system to do, the verification cycle is too long. I was going to do it.” Can’t the product team clearly define the transformation? The operation team visited customers extensively to collect feedback. Advertisers won’t pay? Technical team door-to-door help customers landing. At that time, there was a third-party software called Business tong with a lot of communication clues, Lu Pengjun himself went to talk about the acquisition. … Lu Pengjun helped this new payment model to verify and run in a short period of time with extreme driving force. Now the average daily flow of this model has exceeded 100 million yuan.
Empowerment is not laissez-faire, but ideological consensus is a magic weapon
In 2018, Lu Pengjun, who has a very bright performance in Baidu, decided to accept a new challenge and join The Joyous Times. “After 12 years of advertising, I want to change the track. In 2016 I saw that entertainment queries accounted for 30% of searches and continued to grow, so my gut tells me there’s still a lot of potential in this industry.” Before Joining, the CTO position had been vacant for 4 years in the public annual report of Mirage. The arrival of Lu Pengjun is a welcome welcome for the company, which has 60% r&d personnel. Lu Pengjun sorted out the technology system of Happy Time into three layers: the bottom layer is the file system, operation and maintenance system and various monitoring, the middle layer is the big data recommendation algorithm and computer vision, and the third layer is the end-to-end technology, APP. One of the most core is the algorithm, Huanju just started, is his own specialty, he will put more energy into the building of this plate; However, the APP technology and livestream technology on the end were not involved in much before, so Lu Pengjun authorized the two elderly people to get together. To Lu pengjun’s surprise: “Although they are all teams of more than 1,000 people, Happy Reunion and Phoenix Nest are completely different. Phoenix Nest team is more focused, get-together belongs to the entrepreneurial style. “Startups have to constantly try new directions. When the pressure is greatest, more than a dozen apps are developed simultaneously, and every direction can’t go wrong. It’s really hard.” Mr Lu laments that part of the pressure comes from communicating with non-technical bosses: “Maybe this is a common pain point for many Internet companies. Sometimes when things go wrong, the boss may not understand or be reassured.” Of course, he did not attribute all the problems to communication. “I authorized the front desk and the back office to two veterans, but authorization does not mean delegating power. For a long time, I ignored the control of these two sectors, and found many problems when I looked back. There is nothing wrong with authorization, but the premise is that you have to go first. When the authorized object grows up and the concept is similar to your own, you can fade out. Ideological consensus is very important.”
Postscript:
In 2020, Lu Pengjun chose to join ViPSHOP as a joint CTO.
In his opinion, e-commerce can change the objective world in a more powerful and extensive way and create more real value.
Due to the short time of arrival, the content of the manuscript can only be regretfully ended at the happy time.
Ask whether have the intention of starting a business? Lu Pengjun gave a negative answer, in his view entrepreneurship, partnership, work…… The key is to exert your influence to help the society. Entrepreneurship is not necessarily the best model. “But if I find that I will regret if I do not do something one day, I will definitely choose to start my own business without hesitation, no matter whether I succeed or not.
Tiger does not leave the mountain, people do not leave the road, see you soon.
This article comes from the public account [Kechuangren], sharing the growth experience and methodology of 35 + technical seniors, the community links thousands of CTO/CIO, welcome to exchange ~