Growing startups all reach a point at which they need to be more data driven to better allocate resources and improve customer experiences. To do so, startups leverage the open source machine learning algorithms which require startups to transform production data into science enabled data. This transition often requires a re-architecture of production systems to big data architecture in order to consolidate the data and ensure it is immutable. However, there are ways to create and enable data science without changing the production. Working in parallel, leveraging new startup technologies and python’s open source community, a startup can incrementally provide actionable insights and machine learning. This talk will walk through WeWork’s journey from a system of many bootstrapped applications to applied machine learning and actionable insights, without requiring redesign or interrupting engineering development.
Ahmed started developing his passion for data science through Cancer tracking at Memorial Sloan Kettering. After writing his first algorithm, he was hooked. He studied robotics at Cornell University working under Dr. Mark Campbell to develop AI algorithms for Human Robot interaction. After concluding his research he co-founded Simpoll, the twitter of polling, to bring relationship learning algorithms consumers. After Simpoll, he embarked on a new adventure: writing tracking and discrimination algorithms for missile defense at Raytheon. Recently, Ahmed made his way back to the startup world working for WeWork as a data engineer. He designed, developed and enabled data science systems and is excited to share his journey with you.