Jingwei Chen

Independant Researcher

Research Interests: AI, including Conbinatorial Search, NLP, Agents, Embodied AI

About Me

I'm a passionate AI researcher and engineer with a deep interest in advancing intelligent systems.

During my PhD, I specialized in heuristic search algorithms, exploring efficient methods for solving complex optimization problems in AI planning and decision-making.

At PuduRobotics, I led the development of the company's second-generation Multi-Agent Path Finding (MAPF) project, optimizing navigation and coordination for multi-robot systems in real-world environments. Additionally, I spearheaded a groundbreaking initiative to build an intelligent voice agent from scratch, leveraging Large Language Models (LLMs) to enable natural, context-aware interactions.

These experiences have fueled my growing expertise and enthusiasm in Natural Language Processing (NLP), LLMs, robotics, and embodied AI—fields where I continue to innovate and contribute.

I'm always excited to collaborate on cutting-edge AI projects. Feel free to reach out!

Education

Ph.D. in Computing Science

University of Alberta

2015 - 2022 | Advisor: Prof. Nathan Sturtevant

Ph.D. studies initiated at the University of Denver (2015-2018) and transferred to the University of Alberta to continue research with my advisor.

M.S. in Computer Science

University of Denver

2013 - 2015

B.S. in Computer Science

Shanghai Jiao Tong University

2009 - 2013

Work Experience

MAPF Engineer

PuduRobotics

May 2022 - Sep 2025

Led the design and implementation of the company's second-generation Multi-Agent Path Finding (MAPF) system.
Built the company's LLM-based, multi-scenario intelligent voice agent from scratch.

Publications

A Memory-Bounded Best-First Beam Search and Its Application to Scheduling Halide Programs

Chao Gao, Jingwei Chen, Tianjun Mo, Tarek Sajed, Shaun Jui, Mingyuan Qin, ..., Wei Lu

Proceedings of the International Symposium on Combinatorial Search (SoCS), 15(1), 74-82, July 2022

[PDF]
Necessary and Sufficient Conditions for Avoiding Reopenings in Best First Suboptimal Search with General Bounding Functions

Jingwei Chen, Nathan R. Sturtevant

Proceedings of the AAAI Conference on Artificial Intelligence, 35(5), 3688-3696, May 2021

[PDF]
Conditions for Avoiding Node Re-expansions in Bounded Suboptimal Search

Jingwei Chen, Nathan R. Sturtevant

Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI), 1220-1226, August 2019

[PDF]
Revisiting Suboptimal Search

Jingwei Chen, Nathan Sturtevant, William Doyle, Wheeler Ruml

Proceedings of the International Symposium on Combinatorial Search (SoCS), 10(1), 18-25, 2019

[PDF]
MM: A Bidirectional Search That Is Guaranteed to Meet in the Middle

Robert C. Holte, Ariel Felner, Guni Sharon, Nathan R. Sturtevant and Jingwei Chen

Artificial Intelligence (AIJ), 252, 232-266, 2017

[PDF]
Front-to-End Bidirectional Heuristic Search with Near-Optimal Node Expansions

Jingwei Chen, Robert C. Holte, Sandra Zilles, Nathan R. Sturtevant

Proceedings of the 26th International Joint Conference on Artificial Intelligence (IJCAI), 489-495, August 2017

[PDF]
Sufficient Conditions for Node Expansion in Bidirectional Heuristic Search

Jens Eckerle, Jingwei Chen, Nathan Sturtevant, Sandra Zilles, Robert Holte

Proceedings of the International Conference on Automated Planning and Scheduling (ICAPS), 27, 79-87, June 2017

[PDF]
The Minimal Set of States That Must Be Expanded in a Front-to-End Bidirectional Search

Eshed Shaham, Ariel Felner, Jingwei Chen, Nathan Sturtevant

Proceedings of the International Symposium on Combinatorial Search (SoCS), 8(1), 82-90, 2017

[PDF]
External Memory Bidirectional Search

Nathan R. Sturtevant, Jingwei Chen

International Joint Conference on Artificial Intelligence (IJCAI), 2016

[PDF]

Blog & Resources

Why Faster AI Code Isn’t Faster Software (and How to Fix It)

Discover the new normal in AI-augmented software engineering: Master blueprints, DREM metrics, and human-AI synergy to slash bugs and build faster systems.