Prasad Kawthekar

Filed a provisional patent for a natural language-understanding assistant that connects to all of a user's apps to execute various workflows, such as finding information and performing actions within those apps. This patent formed the foundational basis for Dashworks.

Describes a fully data-driven model that learns to perform a retrosynthetic reaction prediction task, treated as a sequence-to-sequence mapping problem. The end-to-end trained model uses an encoder-decoder architecture with two recurrent neural networks, demonstrating success in solving similar sequence-to-sequence problems.

Focuses on reasoning about the performance of software systems in dynamic environments, particularly under various configurations. The work involves learning a black-box model based on real measurements to understand system behavior and performance.