Mengdi Wang

Mengdi Wang is a professor of electrical and computer engineering at Princeton whose research includes decision making and reinforcement learning. She graduated from Tsinghua University in 2007, completed an MS and Ph.D. at MIT in 2013, and joined Princeton’s operations research faculty by 2018. Her work includes a NeurIPS algorithm for near-optimal Markov decision policies, an mRNA language model that improved protein production in lab tests, and reinforcement-learning methods for wireless chip design. She received the American Automatic Control Council’s 2024 Donald P. Eckman Award.

Now at: Princeton University

Timeline

2007–2013

Tsinghua and MIT degrees

She graduates from Tsinghua University in 2007 and earns an MS and a Ph.D. in electrical engineering and computer science at MIT in 2013.
By 2018

Assistant professor at Princeton

By 2018, she is an assistant professor in Princeton’s Department of Operations Research and Financial Engineering.
2018

Near-optimal MDP policies

In 2018, she coauthors a NeurIPS paper that computes an ε-optimal policy for discounted Markov decision processes from a generative model, with near-optimal time and sample complexity.
2020

Google DeepMind sabbatical

In 2020, she spends a sabbatical year at Google DeepMind studying techniques to improve reinforcement learning.
2024

An mRNA language model

In 2024, she and the biotech startup RVAC Medicines develop a language model that treats biological sequences like text and generates mRNA variants. Lab experiments show a 33% improvement in protein-production efficiency.
2024

A Donald P. Eckman Award

In 2024, the American Automatic Control Council gives her the Donald P. Eckman Award for contributions at the intersection of control, dynamic systems, machine learning and information theory.
June 2025

AI for wireless chip design

In June 2025, she contributes AI and machine-learning research to a Princeton-led wireless-chip project that uses reinforcement learning and RFdiffusion models for automated design.
As of October 2026

Professor at Princeton

As of October 2026, she is a professor of electrical and computer engineering at Princeton, where she studies decision making and reinforcement learning under uncertainty.