Chapter 539: The Gap Between Genius and Talent (First Update)

"Interestingly, it says it's ten bugs, but in reality, the code itself is a big bug. Jon Skeeter thought for a moment, his fingers darting across the keyboard, and lines of code appeared on the screen. This is already the fourth problem, and Jon Skeeter estimates that there will be at least one more programming application problem to follow.

The questions on this website are very interesting, first of all, the candidate's mathematical skills are examined, and the next three questions have their own emphasis, although the correct question in front of you seems simple, but in fact, there are a lot of traps hidden in it, a little careless, you may step into it.

Jon Skeeter showed a smile, this questioner is not bad, but compared with himself, there is still a certain gap, let me teach you, how to write real code.

After clicking confirm, as Jon Skeeter expected, the fifth question was indeed a C++ algorithm programming.

"Handling 5,000 target objects at the same time??" Jon Skeeter looked at the time in the upper right corner of the screen, there were still two hours and forty minutes left, "This seems to be the domain of mainframes and supercomputing, what the hell does this company do?"

No matter how difficult the programming problem is, Jon Skeeter can't be bothered, just a little thought, Jon Skeeter typed the code in the programming environment given on the website.

"That's it. Jon Skeeter looked at the time, and it took him about twenty minutes.

After clicking Confirm, the page changes, "Congratulations on completing all the tests, we will contact you after evaluating your results, please leave your contact information in the space below." ”

Jon Skeeter smiled, his fingers racing, "My contact details are already hidden in my answer, please look carefully." ”

In any field, the best talent has its own pride. Although Jon Skeeter is young, his level is already the top handful on this planet, and everything else is fine, in terms of technology, since you have examined my level, I will also examine whether you are qualified to examine my level.

β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”β€”

So far, 22 computer scientists have joined the AlphaGo team, divided into six working groups to complete the entire set of algorithms. Cowys' traders are assisting our scientists in refining the functionality of the automated trading software, and the development of a beta version is expected to be completed within this week. Gary Harbertton, CEO of AlphaGo, introduced to Zhang Chen.

Gary Harberton is the director of the Intel Computing Center, and was recruited by Zhang Chen to be the CEO of AlphaGo with a high salary, with a strong management background and technical background.

Frankfinn has negotiated with both the New York Stock Exchange and the NASDAQ, agreeing that Cowyce, as a parallel service provider of the two exchanges, will allow Cowys' servers to be hosted in the exchange's computer room and directly connected to the exchange's servers after the official implementation of AST.

The AST will officially take effect on January 1 next year, which means that there are less than four months left for Zhang Chen. During these four months, AlphaGo had to come up with a complete set of automated trading algorithms and hand them over to Cowes of Belrankfin.

AlphaGo is another company founded by Zhang Chen that specializes in algorithms and supercomputers, and specializes in providing technical support to Kouyce.

It has to be said that high-frequency trading is a money-burning industry. That's not going to do much, it's already consumed close to $20 million, and if you want to really make something, you'll need to invest at least another $30 million or so.

However, compared to the high profits of high-frequency trading, these investments are simply negligible.

Even if 20 years later, the operating costs of various high-frequency trading service providers are rising year by year, and under the arms race between each other, it will cost hundreds of millions or even billions of dollars just to upgrade the hardware, but these service providers can still make more than 10% of the annual profit, which shows the high profit of this industry.

But now, the industry has not even started, although there are some automated trading companies, but these companies are not yet in the climate, to complete an order or cancel an action at least a few seconds, with high-frequency trading thousands of trading actions per second is not comparable at all.

"How has the delay problem been solved so far?" Zhang Chen threw out a core question.

Gary Haberton was slightly embarrassed: "Due to hardware limitations, it is still difficult to reduce latency to less than 100 milliseconds. ”

Zhang Chen frowned slightly: "Hardware? Mainframe and optical cable can theoretically control the delay at least at the nanosecond level, how can there be such a big gap?"

Gary Harberton stammered: "This... Belch... Theory is theory, but in fact ... ”

Zhang Chen waved his hand: "A delay of one hundred milliseconds has no meaning in practical application, you just need to tell me how much time it will take to reduce the delay to less than ten milliseconds." ”

Gary Harberton grunted for a long time: "We'll continue to optimize the code, but if we want to reduce the latency to ten milliseconds, it will take at least three months." ”

"Three months?" Zhang Chen frowned, this development time is too long, whether it is lightning trading or high-frequency trading, the most important thing is speed and system stability, it must be debugged for a long time to ensure that the system is foolproof before it is put into use, otherwise once it goes down, it will make the market suffer huge losses. The time given by Hubberton does not allow enough time for debugging, which will inevitably delay the deployment of the Coyce server on major exchanges.

"One month, I can only give you a month, whatever method you use, it must be done within a month. Zhang Chen held out a finger, "With a limit of 10 milliseconds, for every millisecond that the latency is reduced, the development team can get an additional reward of one million dollars." ”

The development team of AlphaGo looked at each other, and the reward policy given by Zhang Chen was not without temptation, but it must also be able to get it.

As an investor, Zhang Chen only wants results, and if Gary Harberton can't solve the problem and come up with the results he needs, he can only replace people.

Haberton has been in the workplace for so many years, even if Zhang Chen didn't say it explicitly, he knew it very well in his heart.

Gary Harberton understands that it's not impossible, but it takes a real team of geniuses to do it, and his current team, while capable of being top-notch, is still far from being talented.

"Gary, there's another one, ah, no, it's two candidates who have solved the puzzle. An AlphaGo researcher said suddenly.

Haberton was in good spirits, and there were so few candidates who could pass the AlphaGo test. The lack of personnel is also a major reason for the slow progress of AlphaGo's research and development. At this time, two candidates can pass the online test, and if they can be poached, at least some of the shortage of manpower can be alleviated.

"Huh?" the researcher pointed to the fourth question, "There's something strange here。。。。。。 He seems to have rewritten this code?"

Haberton also has a deep technical background, and when he looks at it, he can't help but feel numb: "God, how is that possible? It's only 30% of the code?。。。。。。 Wait, what does this code mean? I see!

"Gary, look at the last paragraph. The researchers were also a little excited, and the coder enjoyed seeing a beautiful piece of code no less than seeing Halle Berry striptease in front of him.

"You see that he seems to have reservations about the solution of parallel processing, and the resources mobilized to process 5,000 requests at the same time are only about one-tenth, if you extend it, maybe... ”

Gary Haberton was like being struck by a bolt of lightning, yes, isn't that what he's struggling with right now?

Gary Haberton said eagerly, "His contact information? I'm going to see him now!"

The researcher shrugged: "He didn't leave his contact information, however, he hinted that his contact information was hidden in his answer, hehe, very personal." ”

Gary Harberton immediately ordered, "Find this man out at once! He's a genius, a programming genius!"

Gary Haberton's heart was full of excitement, thinking of this candidate's concise and beautiful code style and genius algorithm setting, perhaps, he can really keep this lucrative job? maybe。。。。。。 Can this prize money really be obtained?

(End of chapter)