Developer Week has many supporters in the technology industry, including top technology companies, including: AWS, Google, Face book, IBM, Red Hat, Microsoft, DJI, Capital One, SAP, Oracle, IBM Watson, Yelp and Squirrel AI Learning. Squirrel AI Learning attended Developer Week 2019, one of the largest developer conference and events in San Francisco. Each year, more than 8,000 developers, engineers, software architects, development teams, managers and executives from more than 70 countries gather in the San Francisco Bay Area to explore and discover the latest developer technologies, languages, platforms and tools.
Developer Week consists of a number of active sections, including the Developer Week 2019 Conference and Expo, hacker marathons, technology recruitment fairs, and a series of seminars, open days and San Francisco city events. In addition, the event will cover both technical and theoretical aspects. The topics to be explored at this year’s conference will include: VR Dev, AI development, blockchain development, IoT development, server-free technology, micro-service technology, etc.
Big Talk, Academic Discussions, Skills Training, Hacker Marathons, Product Show Areas, Corporate Recruitment Desks at the 64,000 square-foot Auckland Convention and Exhibition Center were packed.
Soft power of developers
The theme of the conference was the technology age for developers, and the organizer wrote in a quote:
There are gradual changes in the technique, followed by fundamental innovations. From block chain and artificial intelligence to big data and quantum computing, developer technology represents the fundamental innovation that people have reaped for many years . In the era of DevTech, developers’ technology and tools are now the most disruptive and fundamental technological innovations on the market. When you build tools for developers, you not only need to implement a small incremental case, but you also build platforms, frameworks, and API to enable new Web, mobile Internet and Internet of things innovations.
The Keynotes of this conference also elaborated on the importance of programmer patterns.
Developer Week closed with the final speech by Dr. Dan Bindman, chief data scientist at Squirrel AI Learning. He gave a technical introduction to his research and development Probabilistic Knowledge State (PKS) the model. The PKS model consists of a multi-tiered student capability group containing more than 20,000 model parameters. The model can accurately judge the real-time learning situation of individual and group students and tailor the content of individual learning for each student to suit them.
In conjunction with other existing project-based response theory Item Response Theory (IRT) Comparing with the Knowledge Space Theory (KST) models of Knowledge Space Theory, The PKS model can simply and quickly answer the questions in the learning process, and truly understand each student’s real-time knowledge and knowledge. It does not need a lot of stage tests to master the students’ learning level, just like other models.
Another advantage of the PKS model is that it can capture more data than other models today.” Dan Bindman said in his closing speech, “PKS model compared with IRT and KST-based model, more relevant information can be obtained from the students’ data and the students’ future learning performance can be predicted more accurately.”
The idealized AI education software in Dan Bindman’s mind needs to judge students’ abilities quickly and effectively. This requires high-quality teaching content, continuous analysis to decompose student knowledge maps, and tailor-made content for students based on student knowledge profiles. What Squirrel AI wants to solve through AI adaptive algorithms is the real two pain points in the education industry: the uneven distribution of educational opportunities and the lack of personalized education. Until then, the former can only be solved through MOOC, while the latter can only be handled through tutoring one-to-one tutoring, which fails to meet the expectations of systemic problem solving through technology.
“This is a great time for technology developers,” he said.
Some people say that when the product from the language to the cloud, from the API to the practice of the scene is not short of perfect, the modern era of technology entrepreneurs show towering; some people say that now is the peak time that software application in the industry automation vientiane . some people say that it is too early to mention the golden years, they should take their own way step by step to see after the effect.
The first stage of artificial intelligence is the development of software for automated programming. The second stage is the automation of software in the industry, where the efficiency gains generated by the automation will stimulate the original market. The third stage is the AI will get out of the computer, affecting the outside world.” Ying Weida CEO Huang Renxun said.
Among them, the hacker marathon, or Hackathon, is the largest developer in the United States, the Hackathon Campaign, the event will be attended by six hundred to eight hundred developers, engineers, fifteen to twenty sponsoring companies set challenges and offer prizes and assess winners. Previous sponsors include Intel, DJI, VMware, Tom, etc., as shown in the image below:
The Hack Marathon Scene Is Raging
Slack, a former computer game company is now the largest business communications system in the United States. Slack has hired Gaosheng to prepare for its listing in 2019. which is expected to be worth more than $10 billion. The unicorn in the enterprise services startup market has only been developing for a few years and currently has over 8 million live users and is recognized as the fastest-growing SAAS (to provide software services to enterprises) company in history.
Slack’s CTO Cal Henderson is the author of the bestseller “Building Scalable Websites.” His concept of scalability is known to the programmer startup community. With this philosophy, while early Slack only provided software to a small company with 5 to 15 people, it was highly compatible when the product was scaled up. As the opening guest, he said that the presence of cloud platforms had further expanded the scalability of future programs.
Henderson said, “There will no longer be a distinction between software companies and non-software companies, many of our customers have started to build in-house software systems at the company. Software will be the product base of the future company. Therefore, the entrepreneur must have systematic product technology thinking and macro overall layout of vision.“
CTO David Fullerton of Stack Overflow also emphasized on the rapid iteration of the product. For a programmer community like Stack Flow, he says, the focus is not on creating a perfect site ahead of time, but on iterating products and optimizing operational thinking quickly.
It’s a web site by the coders, for the coders. David Fullerton shares the technology provider’s a question and answer web site with Stack Overflow (developed by the technology provider, for the technology provider’s presence) which has current monthly traffic more than 1 billion views and 17 million questions and answers per month. He said the data represented more than just traffic, and it represented the time that countless developers had saved for the developers to improve the efficiency.
A common question for technology developers is: What is Stack Overflow worth? Nothing. But everything.
In addition to the macro layout, the leadership of tech entrepreneurs is crucial, and Nick Caldwell is the chief product officer for Looker. The subscription service of Looker, which has raised $280 million so far, provides companies with a wide choice of either using a visual dashboard or accessing data from applications such as Slack or Sales force.
Looker’s Chief Product Officer Nick Caldwell emphasizes leadership, mission and passion. Before joining locker, Nick Caldwell served as vice president of engineering for Reddit, responsible for building and operating the fourth-most-visited website in the United States. Prior to joining Reddit, he held various positions in microsoft’s engineering leadership positions, including natural language processing, enterprise search, machine learning, in-memory databases and business intelligence.
He said the focus of technology developers should be on leadership, whether leading the technology team as a whole or working with other departments. While there are different directions of leadership from jobs to Bezos to Musk to Zuckerberg, what technology developers really look for is their mission and passion.
Technology developers, big business companies love you.
Some people focus on the soft power of the technology Developer, and others focus on the imagination space in which the test is applied to their platform. NVIDIA‘s exposure to Developer Week is mainly used in the Developer platform ISAAC for its robotics technology. The representative of NVIDIA was Claire Delaunay, vice president of engineering for NVIDIA, which is in charge of the Isaac Robotics Program.
From the view of a former entrepreneur, Claire Delaunay understands the early pain points that technology entrepreneurs develop in the development of technology. Prior to joining NVIDIA, she was Uber’s engineering director and Google Robotics project leader. But what made her so famous in Silicon Valley was her start-up experience, in 2016, Uber acquired Otto, a startup company she co-founded with less than a year, for a whopping $680 million. Claire Delaunay has also co-founded two other startups, Botiful and Robotics Valley.
Robot entrepreneurs are concerned how to get their products to collect more data,” Claire Delaunay said. “But pretty soon you’re going to find that privacy issues are more stressful for entrepreneurs than data collection.” So, once an entrepreneur, she joined Nvidia to run the ISAAC program. What makes NVIDIA ISAAC robots special is the virtual world, where everything obeys the laws of physics but not the laws of time. According to the report on ISAAC, “In real life, you want to train a machine to learn how to play hockey. You need to put the puck in front of the machine and teach it over and over again. Thousands of times, it takes a lot of time. In the virtual world, however, machines can do this many times a second without following the laws of time in the real world. And you can train a bunch of machines to play at the same time, find the smartest one inside, copy its “brain” program, create a bunch of the same machines, and then continue to train screening.
After introducing the ISAAC platform, Claire Delaunay said, We know the meaning and purpose of robots is to solve the special problem, Nvidia’s ISAAC exists to help entrepreneurs solve the general problem, so that entrepreneurs can better focus on the areas they focus on.
We all thought that the age-old issue of the API would have been solved 10 years ago, “But in fact, yesterday, I also heard users say that complex, off-line buying was a nightmare.” Renaud Visage, a technical architect and co-founder of Eventbrite. In 2006, he created Eventbrite in partnership with Kevin and Julia Hartz, which was founded in 2006, And listed on the New York Stock Exchange in 2018, it has grown into the world’s leading global ticketing and event technology platform.
“A developer’s greatest need for presentation and appeal in a platform product is your creativity. It is also the source of things. With this, we can combine the data with various solutions to create a better KPI and a smoother user experience,” Renaud Visage added.
Although Renaud has a little bit micro-words about the API’s application interface, the application of API has been in the global development of sunshine and efficiency. Overall technology has evolved dramatically over the past few years: HATEOAS allows for runtime parsing in a hypermedia semantic manner, JSON-LD format is increasingly accepted by API providers, and formats such as ALPS allow businesses to provide semantic support for interface data and usage scenarios. Technology in China, because of the vast consumer application scenario, the account access system is pure: from WeChat small programs to QQ big packages. From ant flowers to bank accounts, mobile applications to VR helmet interfaces, the layout of API has blossomed and has produced results everywhere.
The Golden Age of Problem Solving by technology
Although in Developer Week didn’t see BAT, but the presence of Squirrel AI Learning, an AI educational institution, also provided a Chinese element to the tech developer’s heyday.
“Squirrel AI Learning” is an AI adaptive learning engine based on advanced algorithms, with complete intellectual property rights. Its leading R&D members come from Knewton, Realizeit and ALEKS, the three largest AI intellectually adapted educational institutions in the world. In the United States, Squirrel AI Learning works with Sandford research Insitute’s educational department, to develop adaptive education.
From computer mice to robotic surgery, SIRI to cancer therapy, the Stanford Research Institute has created new industries, billions of dollars in market value and enduring social value in more than 70 years of innovation. The Stanford Research Institute is also known as the “Source of Innovation in Silicon Valley” for its outstanding contributions in new materials, engineering, education, the military, information technology, biotechnology, and the energy environment, among other areas.
At Developer Week, Squirrel AI Learning and SRI teamed up on a human-computer war case study. Sam Wang of SRI , the Squirrel AI Learning education researcher, shared with technology developers research on AI’s assessment of AI’s educational products based on the example of human–machine warfare. In four controlled trials in China, the students were randomly and evenly divided into the experimental group that used the Intelligent Adaptation System and the control group of experienced teachers who had extensive teaching experience with Squirrel AI Learning. After a unified pre-learning test, the scores of the two groups of students were recorded, and a final test was conducted after about six hours of study. The results showed that students who used Squirrel AI Learning had an average score of about 6 percent higher than those in the control group.
For thousands of students to cross the single-log bridge in the college entrance examination in China, a one-point difference would be a huge loss, let alone a 6%.
“We’ve found the power of adaptive learning,” Sam Wang said. “Artificial intelligence is simple and doable in education, and it’s quite conceivable.” Squirrel AI Learning based on “human–machine war” series experiments were also included in the AERA, the largest educational conference in the United States.