Here's what we're reading at MongoDB:
Intern Spotlight: Jason Hu
This year, MongoDB welcomed 33 university students to our intern program in Engineering, Marketing, and Education. In this series, we'll introduce you to the talented students who are helping us transform development and operations for how we run applications today. I had the chance to sit down with Jason Hu who spent the summer in our Palo Alto office working with the CAP team! Where do you go to school, what is your major, and what year are you in? I'm a rising senior at Brown University, where I study computer science. What is your role at MongoDB? I’m a software engineering intern on the CAP team. How did you find out about the internship program at MongoDB? Why did you choose to come to MongoDB? I actually started computer science a year late, so I didn’t really know about tech companies, let alone MongoDB until late in college. As I did more CS I learned from upperclassmen and recent graduates about the company, and it seemed like a cool mix of technical challenges that thought big. What’s your hometown? I actually grew up in the Silicon Valley—I’m from Los Altos, California. My friends and I would walk by Facebook when it was still a small startup in Palo Alto. Did you have previous experience using MongoDB before you arrived? If so, how are things different now that you work at MongoDB? If not, how did you learn MongoDB and how was the education process? I had no experience whatsoever. To be honest, databases had been a huge, intimidating block of computer science and software engineering that I haven’t really touched before. I figured throwing myself in the deep end would be the best way to learn, and after one week of orientation I definitely learned more than an entire semester. All the on-boarding staff and mentors were terrific—they should consider becoming professors, after they retire. Working on actual projects has also taught me loads about databases—not just how to use them, but how to think about them in the larger picture of an enterprise or business. Bike or public transportation to work? I drive, actually. One of the perks of being at home, I suppose, is having my old car. What’s a typical day (or week) for you? Well, it’s really hard to say since every week I’ve had has been so different. Some days will be very focused, where I can put on some ear-buds and crank out code. Other days will be back-to-back meetings or presentations, where I’ll be learning about MongoDB’s history or theoretical concepts about databases or road mapping my project for the next week. And I can’t forget the fun! A typical week always has a bunch of great events, from smaller game nights to larger company outings, such as to indoor skydiving, Giants games, or go-karting. The intern coordinators really out-do themselves. What do you love most about MongoDB? Definitely how much I’ve learned. I can say without exaggeration that I learned more about databases during onboarding than an entire semester of the class. But it’s more than just coding chops: Watching the ins-and-outs of a startup has been fascinating. For me, it’s easy at school to be stuck in a narrow path of problem solving—improving runtimes, cleaning code, etc.—but the challenges outside of code are usually not so neat and controlled. Hearing about the problems of marketing, finance, design, and HR has been incredibly illuminating for so much work we take for granted. What’s the most challenging aspect of your job? Adapting and learning on the fly is definitely something I need to work on more. In school, we have a general sense that a right answer exists, and what it should look like. In engineering in the real world, however, there aren’t necessarily the right answers in the back of the book. Now as an engineering intern, my main challenge isn’t finding the right answer, but rather figuring out whether a problem is feasible, what the solution looks like, and whether it’s worth my time. And this isn’t an easy process. During the course of the summer, as I learned and tested different languages and programs, my project had to change and sometimes backtrack as I adapted. What’s do you hope to accomplish while you’re here? The cool thing about Cluster Bingo was that I got the start it from scratch—decisions for structuring the code, as well as library and language decisions, were all mine. So my goal wasn’t necessarily to rush out a fully functional version. Instead I wanted to think about what are its long-term goals, and how to design it in such away that enabled the growth and continuation of the project after I leave. What’s your favorite Seamless lunch order? Definitely any sandwich from the Ace of Sandwiches. Name one secret skill you have, unrelated to work. I’m a really mean baker. Literally. Apparently I get really bossy, but my cakes also are fantastic. Where do you want to be in five years? I picked this question because, well, as a rising college senior it’s kind of on my mind a lot. The honest answer is “I don’t know,” but I can say that one way or another I know I’ll be programming. Which isn’t necessarily to say I’ll be a software engineer: The past few months, I’ve realized that I can use code within any of my interests—advocating social and economic justice, product design and user interfaces, and even my original major of biology and bioengineering. Given how much my interests and skills have changed in the past five years, it’s impossible for me to really know where I’ll be in the upcoming five. All I hope is to somehow find a way to combine some or all of my interests through coding, whether that’s through a software company, start-up, or NGO. Kindle or book? What’s your favorite book? I haven’t read anything on Kindle yet, so I’m going to have to say books. However, the moment Amazon figures out how to give Kindles that old-book smell, I might have to give it a shot. It’s really hard for me to pick a favorite book, but the one that I always find myself rereading is the graphic novel Blankets by Craig Thompson. His illustrations and paneling are gorgeously done, and he has a great ability of capturing little, human interactions in his characters. Every time I read it I find something new. Describe your perfect weekend. Wake up Saturday morning and start off with hiking or the gym. Then I like going up to a museum in San Francisco: The Academy of Sciences in Golden Gate Park never gets old. Afterward, catching up with friends over dinner, before going out for the night or hanging out at their apartments. The next morning usually entails people watching in the park, and finally settling into a book on the train ride home.
Australian Start-Up Ynomia Is Building an IoT Platform to Transform the Construction Industry and its Hostile Environments
The trillion dollar construction industry has not yet experienced the same revolution in technology you might have expected. Low levels of R&D and difficult working environments have led to a lack of innovation and fundamental improvements have been slow. But one Australian start-up is changing that by building an Internet of Things (IoT) platform to harness construction and jobsite data in real time. “Productivity in construction is down there with hunting and fishing as one of the least productive industries per capita in the entire world. It's a space that's ripe for people to come in and really help,” explains Rob Postill , CTO at Ynomia. Ynomia has already been closely involved with many prestigious construction projects, including the residential N06 development in London’s famous 2012 Olympic Village. It was also integral to the construction of the Victoria University Tower in Australia. Link to Podcast Episode Here “These projects involve massive outflow of money: think about glass facades on modern buildings, which can represent 20-30 percent of the overall project cost. They are largely produced in China and can take 12 weeks to get here,” says Postill. “Meanwhile, the plasterer, the plumber, the electrician are all waiting for those glass facades to be put on so it is safe for them to work. If you get it wrong, you can go in the deep red very quickly.” To tackle these longstanding challenges, Ynomia aims to address the lack of connectivity, transparency and data management on construction sites, which has traditionally resulted in the inefficient use of critical personnel, equipment and materials; compressed timelines; and unpredictable cash flows. To optimize productivity, Ynomia offers a simple end-to-end technology solution that creates a Connected Jobsite. Helping teams manage materials, tools, and people across the worksite in real time. IOT in a Hostile Environment The deployment of technology in construction is often fraught with risk. As a result, construction sites are still largely run on paper, such as blueprints, diagrams and models as well as the more traditional invoices and filing. At the same time, there is a constant need to track progress and monitor massive volumes of information across the entire supply chain. Engineers, builders, electricians, plumbers, and all the other associated professionals need to know what they need to do, where they need to be, and when they need to start. “The environment is hostile to technology like GPS, computers, and mobile phone reception because you have a lot of Faraday cages and lots of water and dust,” explains Postill. “You can't have somebody wandering around a construction site with a laptop; it'll get trashed pretty quickly." Enter MongoDB Atlas “On a site, you might be talking about materials, then if you add to that plant & equipment, or bins, or tools etc, you're rapidly getting into thousands and thousands of tags, talking all the time, every day,” said Postill. That means thousands of tags now send millions of readings on Ynomia building sites around the world. All these IoT data packets must be stored efficiently and accurately so Ynomia can reassemble the history of what has happened and track tagged inventory, personnel, and vehicles around the site. Many of the tag events are also safety critical so accuracy is a vital component and packets can't be missed. To address these needs Ynomia was looking for a database that was scalable, flexible, resilient and could easily handle a wide variety of fast-changing sensor data captured from multiple devices. The final component Postill was looking for in a database layer was freedom: a database that didn't lock them into a single cloud platform as they were still in the early stages of assessing cloud partners. The Commonwealth Scientific and Industrial Research Organisation , which Postill had worked with in the past, suggested MongoDB , a general purpose, document-based database built for modern applications. “The most important factor was that the database is event-driven, which I knew would be difficult in the traditional relational model. We deal with millions of tag readings a day, which is a massive wall of data,” said Postill. A Cloud Database Ynomia is using MongoDB Atlas , the global cloud database service, now hosted on Microsoft Azure. Atlas offers best-in-class automation and proven practices that combine availability, scalability, and compliance with the most demanding data security and privacy standards. “When we started we didn't know enough about the problem and we didn't want to be constrained," explained Postill. "MongoDB Atlas gives us a cloud environment that moves with us. It allows us to understand what is happening and make changes to the architecture as we go." Postill says this combination of flexibility and management tooling also allows his developers to focus on business value not undifferentiated code. One example Postill gave was cluster administration: "Cluster administration for a start-up like us is wasted work," he said. "We’re not solving the customer's problem. We're not moving anything on. We’re focusing on the wrong thing. For us to be able to just make that problem go away is huge. Why wouldn’t you?" Atlas also gives Ynomia the option to spin out new clusters seamlessly anywhere in the world. This allows customers to keep data local to their construction site, improving latency and helping solve for regional data regulations. Real Time Analytics The company has also deployed MongoDB Charts, which takes this live data and automatically provides a real time view. Charts is the fastest and easiest way to visualize event data directly from MongoDB in order to act instantly and decisively based on the real-time insights generated by event-driven architecture. It allows Ynomia to share dashboards so all the right people can see what they need to and can collaborate accordingly. “Charts enables us to quickly visualize information without having to build more expensive tools, both internally and externally, to examine our data,” comments Postill. “As a startup, we go through this journey of: what are we doing and how are we doing it? There's a lot of stuff we are finding out along the way on how we slice and re-slice our data using Charts.” A Platform for Future Growth Ynomia is targeting a huge market and is set for ambitious growth in the coming years. How the platform, and its underlying architecture, can continue to scale and evolve will be crucial to enabling that business growth. “We do anything we can to keep things simple,” concluded Postill. “We pick technology partners that save us from spending time we shouldn't spend so we can solve real problems. We pick technologies that roll with the punches and that's MongoDB.” When we started we didn't know enough about the problem and we didn't want to be constrained," explained Postill. "MongoDB Atlas gives us a cloud environment that moves with us. It allows us to understand what is happening and make changes to the architecture as we go. Rob Postill, CTO, Ynomia