I’m an AI researcher currently at OpenAI. My research interests are in universal learning mechanisms that can be used to scalably power foundational multimodal applications. I graduated from UC Berkeley in 2021 where I worked with Pieter Abbeel and Igor Mordatch on reinforcement learning.

Email: kzl at berkeley dot edu

Most recent blogs:

• Mar 2024 — Spending Inference Time
• Feb 2024 — LoRAs as Composable Programs
• Feb 2024 — Unifying RLHF Objectives

Papers:

• Jun 2021 — Decision Transformer
• Mar 2021 — Pretrained Transformers as Universal Computation Engines