About Me
I am a Research Scientist at Meta, on the AI and Systems Co-Design team. I was previously on the MSL Infra team.
My work focuses on high-performance ML infrastructure and AI–systems co-design, such as (1) optimizing post-training and inference for Meta’s Muse Spark models through GPU kernel optimization in Triton, CUDA, and CuteDSL, as well as GPU networking techniques such as device-initiated networking and comm-comp overlap; and (2) building performance models that project the impact of new model architectures, hardware, and inference techniques to guide AI–systems co-design decisions.
I received my PhD in Electrical and Computer Engineering from Purdue University in 2025, advised by Prof. Y. Charlie Hu. My doctoral research focused on efficient ML inference systems.
News
💻 [Aug 2026] Invited to serve on the ACM MobiCom 2027 TPC. Please consider submitting!
💻 [Mar 2026] Invited to serve on the IEEE ICNP 2026 TPC. Please consider submitting!
💻 [Oct 2025] Invited to serve on the MobiSys 2026 TPC. Please consider submitting!
🎉 [Jul 2025] Received USENIX ATC 2025 travel grant - many thanks to USENIX!
🎉 [Jul 2025] PPipe is accepted by USENIX ATC 2025 - see you in Boston, July 7-9!
🎉 [Nov 2024] Paper on 5G Multi-carrier access is accepted by PAM 2025!
Replication: "Performance of Cellular Networks on the Wheels"
ACM Internet Measurement Conference (IMC '25)
PPIPE: Efficient Video Analytics Serving on Heterogeneous GPU Clusters via Pool-Based Pipeline Parallelism
2025 USENIX Annual Technical Conference (ATC '25)
A Large-Scale Study of the Potential of Multi-Carrier Access in the 5G Era
Passive and Active Measurement (PAM '25)
High-Fidelity Cellular Network Control-Plane Traffic Generation without Domain Knowledge
ACM Internet Measurement Conference (IMC '24)
ARISE: High-Capacity AR Offloading Inference Serving via Proactive Scheduling
ACM International Conference on Mobile Systems, Applications, and Services (MobiSys '24)
Performance of Cellular Networks on the Wheels
ACM Internet Measurement Conference (IMC '23)
Can 5G mmWave Enable Edge-Assisted Real-Time Object Detection for Augmented Reality?
International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS '23)
AccuMO: Accuracy-Centric Multitask Offloading in Edge-Assisted Mobile Augmented Reality
Annual International Conference On Mobile Computing And Networking (MobiCom '23)
How to Pipeline Frame Transfer and Server Inference in Edge-assisted AR to Optimize AR Task Accuracy?
International Workshop on Edge Systems, Analytics and Networking (EdgeSys '23)
An in-depth study of uplink performance of 5G mmWave networks
Workshop on 5G and Beyond Network Measurements, Modeling, and Use Cases
Do We Need Sophisticated System Design for Edge-assisted Augmented Reality?
International Workshop on Edge Systems, Analytics and Networking (EdgeSys '22)
Do Larger (More Accurate) Deep Neural Network Models Help in Edge-assisted Augmented Reality?
SIGCOMM Workshop on Network-Application Integration (NAI '21)
Throughput Prediction on 60 GHz Mobile Devices for High-Bandwidth, Latency-Sensitive Applications
Passive and Active Measurement Conference (PAM '21)
Supporting Delay Tolerant Networking: A Comparative Study of Epidemic Routing and NDN
Workshop On Information-Centric Networking Solutions for Real-world Applications
Distributed Dataset Synchronization in Disruptive Networks
IEEE International Conference on Mobile Ad-Hoc and Smart Systems (MASS '19)