Speaker "Lokesh Rajaram" Details Back



Cross product mapping for deriving deep insights


At Intuit, we place data at the heart of every important product decision. With data from 64+ products and 40+ million customers, data is stored in multiple formats each providing unique insights. Here we will see how cross product data is mapped and augmented to derive deep insights. Data sources range from transactions to clickstream to metadata. Data collection and processing is streamlined across Intuit products and services to ingest and process variety of data sources mentioned earlier. Each one of these dataset can unravel an insight but the real potential is by combining them and augmenting them with custom algorithms and building models around the data which can get better over time will unleash information which we call as deep insights. Performing or deriving such insights involve six stages: 1. Understanding the problem, 1. Figuring out the required data sources 3. Building pipelines to bring data together 4. Building models and features to operate on the data 5. Connecting people and products with the derived insights 6. Feedback. Let's take a look at how it is done at Intuit.


Lokesh Rajaram is a Lead Engineer at Inuit involved in building the data platform and pipelines that serves variety of products like TurboTax, Quickbooks, Mint. He likes the challenges of developing and operating large scale distributed applications. He is always excited by the opportunity to create generic solutions and automating them so that it can be leveraged across the products and services with ease. He holds a master’s degree in Computer Science from New Mexico State university.