Hi there, I’m Rashmeet!

I’m a Ph.D. candidate in Computer Science at Arizona State University finishing my dissertation on context-aware abstractions for generalizable long-horizon sequential decision-making. My work sits at the intersection of reinforcement learning, planning, and generative AI, specifically, how agents can learn structured representations that let them reason and act reliably across long-horizon tasks.

I’ve had the chance to work on problems at real scale. At Amazon Air, I designed and trained an attention-based foundation model for large-scale vehicle routing, using LLM-inspired supervised and RL fine-tuning to produce real-time decisions under operational constraints. At LinkedIn, I worked on offline RL for LLM-based task-oriented agents. Before that, I built an AI system for analyzing UV spectra from Hubble Space Telescope using probabilistic logic, work that earned a Chambliss Astronomy Achievement Award from the American Astronomical Society.

More recently I’ve been working on skill learning for LLM agents and contributing to diffusion-guided task and motion planning for robotics. The common thread across all of it is the same question: how do we build agents that generalize in long-horizon settings?

Finishing my PhD in early 2026 and open to Applied Scientist/ML Engineer/Research Scientist roles in GenAI agents, decision-making systems, and embodied AI.