I am a Research Scientist at Adobe Research (AIR Lab) in San Jose, CA. My research spans agentic AI evaluations, benchmarking, and reinforcement learning environments, as well as audio language modeling and full-duplex speech systems. I completed my Ph.D. in Computer Science at the University of Maryland, College Park, advised by Dr. Dinesh Manocha, and an M.S. in Computer Science from UMD (2021).
My work on agentic AI evaluations and reinforcement learning focuses on building realistic benchmarks, interactive environments, and evaluation frameworks for multimodal agents. I am particularly interested in assessing complex, multi-step workflows involving reasoning, tool use, and decision-making, and developing environments that support reinforcement learning-based post-training and continuous improvement of agent capabilities.
My research in audio language modeling and full-duplex speech systems focuses on enabling natural, real-time spoken interactions. I study models that can listen, reason, and speak simultaneously, handle interruptions, backchannels, and conversational turn-taking, and operate beyond conventional turn-based speech pipelines. This builds on my earlier work in speech recognition personalization and neural speech synthesis.
I maintain a full list of my publications on the Research page. I’m always happy to hear about potential research collaborations — feel free to reach out!
Email / CV / Google Scholar / LinkedIn / GitHub