I will complete my PhD in computer science at UC Irvine Winter 2025, co-advised by Sameer Singh and Roy Fox. I research methods for language model agents, applying my expertise in reinforcement learning and natural language processing. In the past, I have published work on instruction following for embodied AI, using LLMs to guide RL exploration, RL for pruning state descriptions, and continual in-context policy improvement. I’m currently working on optimizing LLMs for multi-turn tasks such as dialogue and code generation.
I maintain implementations of LLM agents for Minecraft, NetHack, and text-based game tasks. I also co-created a popular LLM-powered bot called MINDcraft for co-op play in Minecraft. I have industry research experience in AI for games from my time on ML Agents at Unity, AI Dungeon at Latitude, and Riot’s Tech Research team.
I am currently on the job market for industry researcher roles, available early 2025. Click here to download my CV as a PDF.
News
- I am on the job market Fall 2024!
- Our paper Skill Set Optimization: Reinforcing Language Model Behavior via Transferable Skills was accepted to ICML 2024
- I will be interning at Riot Games Summer 2024 on Tech Research’s Generative AI team
- Our paper Selective Perception: Optimizing State Descriptions with Reinforcement Learning for Language Model Actors was accepted to NAACL 2024
- I gave a guest lecture in UCI’s graduate-level RL course on all things RLHF
- I will be interning on AI2’s Aristo team Fall 2023
- Our work Do Embodied Agents Dream of Pixelated Sheep?: Embodied Decision Making using Language Guided World Modelling was accepted at ICML 2023