Paloma Sodhi
paloma.sodhi at gmail dot com
Hello! I’m a Research Scientist at ASAPP, developing advanced ML and NLP technologies for dialog agents that act. My research focuses on LLM agents, reinforcement learning for dialog, and alignment using AI feedback.
In my current role, I work on building LLM agent models that both converse with customers and take actions, such as tool usage and UI interactions. I am excited about applications that help free up humans from mundane tasks, allowing them to be more creative and make better decisions.
Previously, I did my Ph.D. in Robotics at Carnegie Mellon University on combining machine learning with graph optimizers in the loop, where my research was generously supported by Meta AI and CMU Presidential Fellowship.