Diagnosing Component-Level Failures in Computer-Use Agents
COLM 2026 (In review)Tianchen Guan, Daisy Xinlei Lin, Xiangjun Wang, Shuyan Zhou
Daisy Xinlei Lin
AI Researcher @ Amazon AGI SF Lab
I build AI agents that act like humans do, because the only thing more fun than being human is trying to recreate parts of it in silicon. I earned my PhD at NYU, where I studied planning and reinforcement learning: how minds and machines decide what to do next when the world is messy, uncertain, and adversarial. These days I work on LLM post-training, reasoning, and agents — systems that don't just talk, but act.
Featured Work
Industry
Amazon Nova Act
ProductionOwned the end-to-end RL training recipe for Nova Act. Designed and shipped an online data curator that improved RL sample efficiency. Diagnosed and resolved a training/inference mismatch that stabilized RL training at scale across agentic tasks. Drove measurable gains in training data diversity and reliability.
Reinforcement Learning Computer Use LLM Post-Training
Read AWS Blog Post →
Selected Publications
Limitations in Planning Ability in AlphaZero
NeurIPS 2024Daisy Xinlei Lin, Brenden Lake, Wei Ji Ma
Paper →Learning How Humans Learn to Play Board Games with GPT-4
AAAI Workshop 2024 · FCAI AI Day 2023Veronica Yeom-Song, Xinlei Lin, Ionatan Kuperwajs, Heiko Schütt, Wei Ji Ma, Luigi Acerbi
Comparing Machine and Human Learning in a Planning Task
CogSci 2022Zheyang Sam Zheng*, Xinlei Daisy Lin*, Jake Topping*, Wei Ji Ma
Paper →Human Planning in Stochastic Environments
PsyArXiv 2025Jordan Lei, Jeroen Olieslagers, Nastaran Arfaei, Xinlei Lin, Wei Ji Ma
Paper →Complementary cortical and thalamic contributions to cell type–specific striatal activity dynamics during movement
Science Advances 2026Enida Gjoni, Ram Dyuthi Sristi, Haixin Liu, Shahar Dror, Xinlei Lin, Keelin O'Neil, Oscar M. Arroyo, Sun Woo Hong, Hannah Kim, Jeffrey Liu, Sonja Blumenstock, Byungkook Lim, Gal Mishne, Takaki Komiyama
Paper →About
When I'm not teaching models what to do, I'm usually snowboarding, hiking, baking desserts, or being supervised by cats who believe they are my bosses.
I moved from NYC to SF to join the RL research team at AGI SF Lab.