Girish Dilip Patil

In reinforcement learning (RL), an RL agent learns in an interactive environment by trial and error using feedback from its own actions, and can make sophisticated multi-step decisions. As a result, RL has broad applicability in robotics, industrial control, finance, HVAC, dialog systems, online advertising, and more.

With global business, there is an ever-growing need to be able to implement multi-region active-active architecture. However, this requires first-order thinking and attention not just to app and database design but also to DNS, monitoring, traffic shaping, and so on.