Fortune picks 20 top talent in the big data industry

Author: Andrew Nusca,robert hackett,shalene Gupta

Translator: Pak

Excerpt from: Fortune Chinese web

Big data is not just about dealing with a lot of numbers, it's going to have to build models, dig deeper, and look for information that might change the way companies operate. I would like to introduce you to the top 20 large data fields.

Pinterest data scientist Andrea Berbink

Pinterest, a graphic social network, data scientist Andrea Burbank is mainly responsible for the company's A/b test to assess the impact on its 60 million of global users of the company's Web site, app's appearance or function changes. If a small module on the Pinterest Web site makes you want to invite friends to play, or some of the recommended emails that it sends to focus on more topics, it's probably the "invisible hand" of the Burbank team that works. "To provide services to billions of of users in the future, we have done hundreds of trials for millions of users," she said at an industry conference this year. "The tests also extend to Pinterest's own operations," he said. One of the biggest achievements of Burbank is to empower other employees of Pinterest so that they can carry out tests as well. She said: "There was only one point of failure, but there is only one point of knowledge, now different." "--andrew Nusca

0xdata company physicist, hacker Annot Canders

Annot Canders grew up in an environment with a strong scientific atmosphere. He was born in Switzerland, a small village called the lower Cassiterite, located between Paul Rosier College (Paul Scherrer Cato), which owns Europe's top particle accelerator laboratory, and the Federal Polytechnic of Zurich (ETH Zürich), the most prestigious technology school in Europe. During the study of particle physics and supercomputers, Canders had built a model of the universe on a computer. He later moved to California, where he worked at the National Accelerated Laboratory at the Center for Linear Accelerator at Stanford University (SLAC Nation Accelerator Laboratory), and then started a business as a founding engineer for Skytree. and several sets of high-performance machine learning algorithms are designed. Now he is the core developer of the Oxdata company's made data analysis platform, the made platform he developed as the best open source Java machine Learning Project by members of the programming community GitHub, and it can also be compatible with popular statistical programming languages--r languages. His title at the Oxdata Company was "physicist and hacker". --robert Hackett

Hortonworks co-founder Alain Meusy

Alain Meusy, the first to work at Yahoo, is now a very versatile source of open-source storage and processing software Hadoop, which was still in its early prototype stage. The task of Merseyside's team is to expand the functionality of Hadoop and use it for Yahoo's web search. So Merseyside developed a resource-load management system called yarn, which is roughly equivalent to one of Hadoop's operating systems. "The first version of Hadoop looks like windows that are still in Notepad, but what we really want is a PowerPoint, Word, and Excel windows," says Merseyside. "That's what yarn's charm is: it allows users to insert multiple applications into Hadoop to store various types of information." "I have two children at home, but yarn is like my third child," says Merseyside. "--robert Hackett

NuoDB CEO Barry Morris

Many technology companies want to launch a technological revolution, but few have the support of the last wave of technological revolution leaders. The NuoDB Company in Cambridge, Massachusetts, has netted the last wave of the "four Kings" of the Database Revolution,--ingres Morgan Taylor, former CEO of the company, and former CEO Mitchell Kezman of Sybase, former CEO of Informix company Roger Spour. Only the Last "king", the CEO of Oracle, abstained, but he is also in the investor list. Why does such an obscure company have such a great appeal? Because NuoDB's technology solves a problem that has been recognized by the industry as the "Holy Grail" for years: How to make a database run on multiple servers. "The key is to use more machines, not bigger machines." "That question sounds simple, but it hasn't been solved," said the company's CEO, Barry Morris. "Duodb Company was founded in 2010, and now the company has signed a major customer-the second largest software manufacturer in Europe (Dassault Systèmes). In addition, Morris said, the company is moving rapidly towards a "new focus". He thinks Duodb will take the center of the stage. "Its significance is not in the size or speed of data, but in the form of a data-oriented model." The ability to continuously improve is its revolutionary. "--andrew Nusca

Beats Music Company large data Engineering director Blaine Rogski

It's not easy to get Blaine Rogski to talk about the details of his work at Beats Music company, but Beats Music has just been acquired by Apple, and it's not surprising that Apple has always had a mystical style. (The deal was announced in May this year and completed in Friday.) But Rogski is still three words of the bank. He says companies are now interested in more in-depth real-time data analysis, and are interested in increasing data sharing within the enterprise. Many other companies also want to use big data to improve the app's personalized experience. So how Rogski going to keep up with the beats company? "You have to understand the secret," he says. Because of my present role, I can not say too much to this question, now let's talk about the big trend. "--robert Hackett

Yahoo Lab researcher Danieres Caisia

As a child, Danieres Caisia the most wanted to be a policeman, for which he also bought a toy motorcycle. But now at Yahoo Labs, his knowledge of the city is absolutely nuanced-albeit at a digital level. Caisia has a PhD in computer science and a postdoctoral degree in urban research at MIT. At present he mainly carries on the wide area city research. For example, he built a game that asked people what kind of urban landscape they liked, and then put the score on Facebook for viral transmission. Caisia will study the results to determine what kind of urban landscape people like or dislike, in order to eventually design better and more beautiful cities. "Computer science is a learning tool for building tools," he said. I want to do something new that can affect the reality. More than half of the world's population lives in cities. "--shalene Gupta

Dru Pavis, head of research group on Computational Ecology and Environment, Microsoft Research

There is a reservation in Microsoft (Microsoft) called Bill Reviews, where employees can present their ideas before the founder and chairman Bill Gates. When Steve Emmitt, head of computer science at Microsoft Research's Cambridge Labs, said in a "Bill review" that a research group on ecology would be funded by the Cambridge Laboratory, " It was a very bad meeting. University Purvis, a Princeton ecologist at Princeton University, said. But in the end Bill Gates changed his mind, and soon Microsoft hired Purvis as the leader of the group to build a predictive model for the Earth's ecosystem. Since the founding of Microsoft's Blue Sky Research department, Purvis has led a team of researchers to develop a "Mattingley model" that simulates all life on Earth. This project is ambitious, although some "Don Quixote" color, but some of its results may eventually be applied to the actual. "Everything that happens in the national economy can be considered in some environmental factors," says Purvis. He casually points to the biggest challenges facing the world, including population ageing, cancer, food security, climate change and alternative energy. For ecology and Biology, Purvis adds: "These things will be the core drivers of the 21st century economy." ”

IBM Watson and Cognitive Cooking Group Senior software engineer Florian Piner

Since IBM's "smart" computer system Watson has defeated human rivals in the jeopardy of television intelligence programs, IBM wants to try how far it can push the boundaries of intelligent computing. IBM recently decided to let Watson enter the culinary world. As a professionally trained chef, Florian Piner is also part of IBM's "Cognitive Cooking" team. "We focus on food because food is something that everyone cares about, and we can easily create prototypes," says Piner. Now I am overjoyed to be able to combine the passion for food and computer science. "The team started with a variety of ingredients and spices, but the combination of food they developed was exponentially growing, creating a vast array of potential gourmet recipes." At this year's SXSW conference in Austin, Texas, for example, Watson's system invented an Australian chocolate burrito in the IBM Gourmet Truck, which includes chocolate, beef, green beans and soybeans. It sounds like a very difficult dish to eat, but Piner says: "It's a great success and it's very tasty." So our goal is to inspire users and help them develop food recipes that they never imagined. "--robert Hackett

Cloudera, chief scientist of the company, assistant professor of genetics and Genomics, Mount Sinai Hospital, Hammerbache

As an assistant professor of genetics and genomics at Mount Sinai Shetty, Jeff Hammerbache, though he does not have a ph. D., has a wealth of experience that makes him absolutely competent. He was one of Facebook's best data scientists, and later abandoned the social media industry, which was cloudera by advertising, and became the chief scientist of the company. Cloudera is an enterprise software vendor based on the Apache Hadoop platform. Now Hammerbache is working with Eric Saudi, the head of genetics and Genomics at Mount Sinai Hospital, to try to bring large data analysis into the health care sector. "He is defining a new field that will later become a science that will give a PhD," the Saudis said. Now there is no discipline to teach him what he has done. "Hammerbache is building the infrastructure needed to manage and compute human health data to build better medical prediction models," he said. "For the healthcare industry and the healthcare sector, now is the time for them to start using the Big data analysis engine," said the Saudis. Our cooperation with people like Jeff is a good illustration of the future of medicine. "--robert Hackett

CloudFlare co-founder Michel Zate

If the internet also has its own CIA, the sign may fall to the CloudFlare company. The San Francisco company monitors 5% of the world's network traffic. One of the company's founders, Michel Zate, met two other partners at Harvard University (Harvard University) in 2009-Mashu Prince and Lee Holloway-and founded the company in the same year. CloudFlare plays the role of a buffer between a website and a malicious user. If CloudFlare determines that a user is a friendly user, it speeds up the services he enjoys. If it determines that a user is a spammer or a robot program, it lowers the speed of his service or asks the other to enter a CAPTCHA. "10 years ago, I knew I wanted to be part of a big, important team, but I didn't know exactly what that meant," Zatlin said. But now I feel lucky because I found CloudFlare. I can't imagine anything better than helping clients run their business better. "--shalene Gupta

Jawbone Data Vice President Monica Rogatti

In the Jawbone Company, Monica Rogatti has two responsibilities. The first is to figure out what the data collected by the company's up smart watch represents. Second, build new products that use this data intelligently. "We found a lot of new information about how we sleep, move and eat, all of which were not discovered before," says Rogati. We used to sleep on 100 people, but now we can study the sleep information of 100,000 people. "That is to say, Rogati and her team can understand why people lose sleep." People in Washington, for example, may be thrilled by a presidential inauguration; residents of Istanbul may not be able to sleep because of protests in the Middle East, and Catholic Rhode Island residents may lose sleep over the pope's resignation-and then the company can feed the information to the Up smart watch and adjust their behavior. "We take full advantage of the insights that these data reflect and use them to encourage people to stay at their best levels," Rogati said. "--andrew Nusca

Euno Zoit, senior researcher at Xerox European Research Center

As a child, Euno Zoit dreamed of becoming a Lego designer. When he was 8 years old, he had the first computer in his life, and it ignited a boy's interest in AI. Now Zoit has entered the Xerox European Laboratory (Xerox Labs Europe), whose focus is on easing traffic jams in Los Angeles. "We don't know much about parking, because it takes a lot of time to observe," says Zoit. "To this end, Zoit's team installed a large number of sensors throughout the city's parking lots. The information collected by the sensor is fed back to a smartphone application so that the driver can keep abreast of which car parks are full and which parking spaces are empty. Even better, the city can adjust the parking fees of these parking lots according to the data to reduce the traffic flow in the highly congested areas. Since the launch of the project in 2012, traffic jams in Los Angeles have eased 10%. --shalene Gupta

Eventbrite, vice president of engineering, Patrick Purs

Patrick Purs once said goodbye to the technology industry for five years, during which time he switched to being a professional poker player. But the professional poker market shrank very badly in 2010, so he decided to return to the technology industry. He did not regret his choice, saying: "The analysis of data is like playing poker." You play thousands of hands, you know people, you work with data, and you look for something conspicuous. These are common. "In the Eventbrite network ticketing company, pools and his team have a recommendation system that allows them to recommend large events that they may be interested in based on what they have already browsed." The system is responding well, with about 1 million people now buying tickets on Eventbrite every week, nearly half of which are repeat customers. Pools's next project is to figure out how best to sell tickets for activities that require reservations.

Dropbox chief Scientist Silvanos Li Hu

Silvanos Li Hu, a "prodigy", took only two years to obtain a double degree in computer Science and mathematics from Stanford University. He entered the financial industry directly after graduating, but the temptation of the technology industry was too strong, so he joined Dropbox in 2012. With his business background and scientific knowledge, he set up a team dedicated to data science research. He used to be in charge of a communications software called Project Harmony that allows Dropbox users to view changes to shared documents in real time and discuss them with them. Another item related to sales is to see if any employees from the same company are using Dropbox at the same time, and if so, provide them with a premium package. Silvanos Lee's boss, Wang, said: "His background is very rare, spanning technology, mathematics and business practice." This makes his data science a standout. "--shalene Gupta

Airbnb software engineer Surabi Gupta

Surabi Gupta is very fond of traveling, often planning travel routes for relatives and friends. As a graduate student at Stanford University (Stanford University) majoring in computer science, she was fascinated by the art of abstracts-just a few words to do it without having to read the whole text. Earlier Gupta had studied rental services company Airbnb during Google's work. She was interested in the possibilities that Airbnb's data offered, so she contacted Airbnb to move to the company. Just four months later, she has significantly improved the Airbnb search engine. Now Gupta is trying to compress all of Airbnb's lists and create summaries so that users can quickly learn about the different styles of the city. "The overall goal is, when people travel, how do we attract them to Airbnb?" she says. How do we let them get our message when they want to travel? "--shalene Gupta

Swati Singh, vice president of American Express GMs im platform and large data project

Swati Singh's technical background is almost unassailable, including a PhD in machine learning from Duke University. But her character is keen on business. She is the main think-tank behind American Express's Myoffers service, a service designed to allow American Express members to get what they want when they need it Anglo. For example, it's noon, and you just want to eat Mexican food, and American Express will send you a coupon for a nearby Mexican restaurant. She is responsible for another tool that allows businesses to compare their annual performance. Her boss, Sestri Durvasura, said: "She has a strong technical background, but she can also talk to other leaders like a leader." She was present every time we discussed the data. "--shalene Gupta

Adobe Digital Index chief analyst Tamara Gavny

Tamara Gavny uses data to predict the future. At Adobe Digital Systems, her team is responsible for predicting problems with data from users who use Adobe cloud services, such as which film will win an Oscar, which blockbuster will be hit, and how many people will shop online during the winter shopping season. Her predictions were really remarkable, and in the Christmas shopping season last year, the Gaffney team forecast was only 1% lower than the actual situation. Her manager, Julie McEntee, said: "Her understanding of technology, coupled with her interest in human behavior, makes her very different." She has a strong curiosity, likes to classify patterns of data, track clues, and make assumptions about the evolution of things. "The next project of Gaffney is to predict all kinds of mobile shopping applications." --shalene Gupta

Rent The runway Chief Analysis Officer Vijai Suberamanian

At first glance, in a company that operates a fashion business, Suberamanian's job looks more than just fresh. But as a fashion-leasing service rent The runway's chief analyst, few people understand the feminine fashion trend more deeply than he does. Shortly after joining the company in 2010, he built an evaluation model that could estimate the missing demand, product life, and the use of inventory clothing. The Viggia model can be said to have saved a lot of money for the company, as the company buys fashion jewelry from fashion designers to rent to customers every quarter. "If you put three data sources together, you can build a model framework that embodies shopping habits," he says. It tells us what kind of clothes we should be looking for, which is the most likely way to get us out of the celebrity van. "-and, of course, try to avoid it. Viggia's next project is a new type of data that integrates unlimited feedback--unlimited is a project launched by the company in an effort to move into everyday casual wear. "Our classic models revolve around the big situations you're going to," says Viggia. You may dress more avantgarde, but if you attend formal occasions, your own style of dressing is not important, it is important to adapt to the occasion. And unlimited is the beginning of the understanding of the user's style of dressing. "--andrew Nusca

Sharethis, vice president of data Science, Quian

Many people may find it hard to believe that Quian's work has affected 95% of American readers. Quian has developed a sharethis called the Social Quality Index (Social Quality Index) at the social networking site, which measures social activities around certain network content and helps advertisers and publishers to target groups. Quian said: "The technical aspects are not difficult, it is difficult to find a business problem, and then apply the technology." Quian a Ph. D. in natural language processing at Carnegie Mellon University (Carnegie Mellon University). Before entering Sharethis, she also led the Advance research team of AOL's advertising.com. Now she likes to meet the challenges posed by mobile phones and does not allow websites to implant cookies to confirm repeated visitors. These require Quian to collect more data. --shalene Gupta

Data Collective company management partner Zakari Bog

As early as 2011, Zakari Bog and Matt Ocdor set up a VC company in the San Francisco Bay Area to focus on entrepreneurship in the Big data field, a vision that was certainly extraordinary at the time. Since then, the company's business has been rising with the wave of big data. "The cost curve was quickly crossed," says Borg. The rapid decline in costs has allowed these new methods to hit the old industries. "This year, the company launched its own third fund, bringing its total financing to nearly 250 million dollars." The company has staked its bets on lendup, a short-term lending start-up, and new companies such as the Memory database service Memsql. It is said that Memsql is much cheaper and faster than the existing memory database available on the market. "Technology opens up unprecedented investment opportunities for the vast market and industry," says Berger. One of our views is that it is exciting that every sector and industry will be completely disrupted by technology. "--andrew Nusca (Fortune Chinese Network)

(Responsible editor: Mengyishan)