A mobile application is only as good as our design and how customers use it. But how do they use it? We’ve got over 35 million devices running our mobile banking platform, and we need to understand each and every one of them. Is the customer enjoying their experience, are they lost, or are they a fraudulent hacker 3000 miles away?
We developed an algorithm to examine the user’s workflow so we can perform near real-time analysis of their online activities. We leverage Spark’s Structured Streaming, ML Pipelines & GraphFrames, and good old fashioned grit to gain insights that allow us to improve our mobile app. For good measure, we added fraud detection to the mix so we can use artificial intelligence to detect any strange or alarming patterns.
Session hashtag: #DevSAIS18
Aaron is a Sr. Director of Data Engineering who has built multiple data lakes and Data Warehouses for major companies across the financial space. He has spent the last couple of years working passionately inside evolving technologies such as Lakehouse, Data mesh, and scalable data system. Aaron holds a master’s degree in information technology from University of Wisconsin.
Kevin Mellott is a Team Lead and Spark Data Developer at FIS, working to improve the data processing pipeline of one of the world's largest FinTech companies. Although he began as a traditional Software Engineer, his career ventured into the world of data science several years ago. Recent projects have included the use of Spark's machine learning pipelines as well as the Python Natural Language Toolkit. When he isn't wrestling with Big Data, Kevin can be found at the local movie theater or ice hockey rink.