Lily Zhang

I’m Lily Zhang, a Tech Lead and Research Scientist in the Bay Area, California. My research focus is: post-training and AI training data. I’ve published at CVPR, ICCV, NeurIPS, RAL [1] [2] [3] [4] [5], and gave a keynote at NVIDIA GTC.

At Latitude AI, I build LLMs, multimodal LLMs, and AI agents for physical AI (Modal GTC panel on agentic post-training). Previously, I led applied R&D at Ford Greenfield Labs (2019–2023) on building perception models and AI training data (IROS keynote).

My latest research (colab w AliCloud) reframes feature engineering as agentic code generation, SFT + RL post-trained AI-infra agent deployed at Alibaba Cloud with 91% adoption (poster at NeurIPS 2025 VLM4RWD, oral at DASFAA 2026). Publishing as Xianling Zhang.

X / CV / In / Email / Scholar