Weiyang "Frank" Wang

About Me

I’m a final-year Ph.D. student at MIT CSAIL’s Network and Mobile Systems (NMS) research group, working with Professor Manya Ghobadi. Before MIT, I earned my B.S. in computer science (honors), advised by Professor Alex C. Snoeren, and my B.S. in physics from UC San Diego. I design adaptive network systems for machine learning workloads, focusing on network architectures, host-adaptation of reconfigurable networks, and programmable datapaths. When I’m not in the lab, I’m usually making coffeeMy favorite coffee beans are natural-processed Ethiopian, and my favorite brew method currently is the Aeropress.
I’m unofficially maintaining the espresso setup on MIT Stata G9—DO NOT mess up the grind settings!
My labmate Jeremy Carin maintains an excellent list of coffee shops in Boston and Cambridge, check it out!, taking or editing photosNothing beats a Ricoh GR IIIx!
I’d like to set up a proper portfolio of photographs in the future.
Remotely related to photography: a physicist really wants to tell you about moiré patterns :) , or lifting heavy things at the gymCurrent PR: Bench press: 225 lbs Squat: 310 lbs Deadlift: 425 lbs.

I am currently on the faculty job market for positions starting in 2026-2027.
[Research Statement] [Teaching Statement]

Research Interests

• Computer Network: Reconfigurable Network, Datacenter Network, Protocol Design
• Systems for Machine Learning: Collective Communication, Performance Modeling, Fault Tolerance

Publications

Posters

• Efficient Direct-Connect Topologies for Collective Communications
  Liangyu Zhao, Siddharth Pal, Tapan Chugh, Weiyang Wang, Jason Fantl, Prithwish Basu, Joud Khoury, Arvind Krishnamurthy
  Poster Session at the 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI), Philadelphia, PA, April 2024

• Zero Buffer Optical Packet Switching Data Center Network
  Shawn Shuoshuo Chen, Weiyang Wang, Manya Ghobadi, Srinivasan Seshan, Peter Steenkiste
  Poster Session at the 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI), Santa Clara, CA, April 2025

Patent

• In-network Optical Inference
  Manya Ghobadi, Zhizhen Zhong, Weiyang Wang, Liane Sarah Beland Bernstein, Alexander Sludds, Ryan Hamerly, Dirk Robert Englund
  US Patent Application Number 18561985

Teaching Experiences

• 6.5820 Computer Networks, Fall 2024 — Guest Lecturer, MIT EECS Department
  Delivered a full lecture on distributed machine learning
• 6.5820 Computer Networks, Fall 2023 — Guest Lecturer, MIT EECS Department
  Delivered part of a lecture on networks for machine learning
• 6.5820 Computer Networks, Fall 2022 — Teaching Assistant, MIT EECS Department
• CSE 8B Introduction to Programming II, Winter 2020 — Tutor, UCSD CSE Department
• CSE 132A Database Principles, Fall 2018 – Fall 2019 — Tutor, UCSD CSE Department

Honors and Awards

• Finalist, Meta PhD Research Fellowship (Computer Networks) — 2022
• MIT Presidential Fellowship — Sep 2020
• UCSD Jacobs School Awards of Excellence (Computer Science and Engineering) — 2020
• CRA Outstanding Undergraduate Researcher Awards, Runners-up — 2019
• UCSD CSE Computer Networking Espresso Prize, Winner — 2018
• UCSD Provost Honors — 2016 to 2020

Professional Services

• Reviewer, IEEE/ACM Transactions on Networking (ToN) — 2025
• Reviewer, IEEE Open Journal of the Solid-State Circuits Society (OJ-SSCS) — 2025
• Reviewer, Elsevier Computer Networks (COMNET) — 2025
• External Reviewer, IEEE International Conference on Network Protocols (ICNP) — 2025
• External Reviewer, ACM Asia-Pacific Workshop on Networking (APNet) — 2025
• External Reviewer, IEEE Global Communications Conference (GLOBECOM) — 2025

Other Services

• Member, MIT EECS REFS
  Served as peer mediator supporting the graduate community and acting as a first point of contact for stress-related issues
• Web Chair, MIT Sidney-Pacific Student Government
  Maintained and developed the administration system for MIT’s Sidney-Pacific graduate housing
• System Administrator, MIT Network and Mobile Systems Research Group
  Constructed and managed a 24-node GPU cluster with electrical and optical networking from scratch

Powered by Jekyll and Minimal Light theme.