About

Umang Moorarka is a highly skilled Machine Learning Engineer III at Adobe with over 5.70 years of relevant experience in the field. He holds a Bachelor of Science (B.S.) in Mathematics & Scientific Computing from the prestigious Indian Institute of Technology, Kanpur. Umang's expertise lies in software development, machine learning, reinforcement learning, and research science. He has worked extensively on building conversion models for creative optimization, incorporating creative features into production models, and generating user-content affinity using network embedding. He has also built a fuzzy ensemble that could beat Bine and FFM at different stages. At Adobe, Umang has worked on multiple projects and has made significant contributions to the company's success. He formulated the problem of Dynamic Creative Optimization as a Contextual Multi-armed Bandit problem and benchmarked multiple Reinforcement Learning based solutions like Thompson Sampling and Upper Confidence bound. He also experimented and implemented variants of Factorization machine-based algorithms with research on explore-exploit trade-off and unbiased offline evaluation. Umang's exceptional skills and knowledge have made him a valuable asset to Adobe. He has mentored interns and worked collaboratively with his team to achieve their goals. His past experiences as a Machine Learning Engineer II at Adobe and his current role as a Machine Learning Engineer III demonstrate his consistent growth and success in the field. Overall, Umang Moorarka is a highly motivated and skilled professional with a passion for machine learning and software development. His academic background, coupled with his extensive experience, makes him an expert in his field. He is always eager to learn and grow, making him an excellent addition to any team.