Jun Qian and Machine Learning in the Real World
Jun Qian’s typical day consists of thinking through real-world machine learning applications, hiring brilliant engineers, and digging through incomparable types of data. Well, that is when he’s not traveling between the Sift Science Seattle and San Francisco offices.
What began as a 20-year friendship with Matt Green – Sift’s Seattle Site Manager and Head of Engineering – has now blossomed into a partnership to help Sift scale its team, develop amazing new offerings for customers, and build better models every day. Jun joined Sift Science in November as the company’s Director of Engineering, and we wanted to get to know him a little better.
“Sift = big data platform + large-scale machine learning + potentially deep learning…plus we’re fighting bad guys. When you combine our knowledge with our purpose, Sift can’t go wrong.”
What about Sift’s machine learning is so appealing?
What’s most inspiring to me about Sift lies in the machine learning challenges. Here, we’re applying machine learning and analyzing data that is far outside the realm of academic algorithms. I think many engineers dream about seeing their work make an impact; at Sift, the opportunity to solve ML problems with data and explore deep learning applications brings that dream within reach.
“I see the opportunity for engineers to have a huge impact solving unique problems.”
At global players like Amazon and Microsoft, fraud is a different problem space. For them, data is contained to each customer, since Amazon and Microsoft just provide the platform or the research. Sift is truly unique; here, companies are compelled to join our global network when they see competing logos. After all, if one of them can get ahead of fraud or train the model to intercept it, that means that the models can work for anyone with a similar use case. I haven’t seen any other competitors in the space that can compare. Sift provides a unique solution as a platform offering real products that benefit customers.
What’s on your mind as you look toward 2019?
We’re ready to scale, we’re just looking for the right people to join our team and help us do it. Our sales and marketing teams are robust, and now we need to ensure that our eng team can support our customers with ever-more stable and scalable systems. Our product is already great and we have an amazing customer base, but I’m focused on how we can continue to do better and serve the best ML possible. I want to help our customers to understand their data, how to analyze it, and what to do with it in order to make their businesses even stronger. We built the sturdy foundation, and now we’re looking toward a bigger scope and better products for our customers.
And what’s it like working at Sift Science?
The Sift Science team is an excellent startup to join. Although we’re a startup, we have mature organizational structure and processes. Every week, there are events and trainings, opportunities to build communities within Sift. I appreciate how much this company invests in encouraging people to learn, converse cross-functionally, and grow. In my few weeks here, there have been several communications courses, fireside chats with other founders and startup leaders, and networking events. Most importantly, however, is the real work that the team is doing. I really like that the people here all care deeply about solving the challenges that customers face, and I’m excited to join them on the front lines in the fight against fraud.
“At Sift, you can apply ML models to a real problem, where you can learn not only fundamental but also practical ML. Where else can engineers learn and practice on sometimes unbelievable real world scenarios?”