Last updated on Dec.16 08:40 UTC+0 2024, New York

About

Hi there! I am a first-year Ph.D. student in Computer Science at NYU Courant, advised by Prof. Aurojit Panda. Before that, I worked as a full-time Machine Learning Platform R&D Engineer in the Applied Machine Learning(AML) department at ByteDance, where I touched Kubernetes/Prometheus ecosystem, industrial cloud infrastructure and large-scale model training. I obtained my BSc in Computer Science at New York University Shanghai, and was advised by Prof. Grace Liu.

Currently my works cover 1) computational systems on CXL disaggregated memory, 2) policy compliance of database-backed systems, and 3) fault tolerance infrastructure for microservice systems. Before that, I have been working on trustworthy network outsourcing for a while with Prof. Grace Liu. They gift me the chance to (seriously) play with Ray, Datafusion, Envoy/Istio, GNS3 Emulator, etc.

I envision my research interest as:

• building easy-to-use and extensible cross-layer acceleration techniques and interfaces (e.g. CXL, smartNIC/DPU, eBPF, and P4 switches) for next-generation cloud systems,

• building efficient and realible cloud clusters through efficient fault tolerance (Analogy: fast backup, sync recovery for programs; better congestion control and multi-path switching for networks) and effective cluster management (Analogy: verified control loop for programs; self-healing and self-tuning for infrastructures).

• improving the observability of cloud systems and networks (e.g. distributed tracing, shared logging, event infrastructure, hybrid timeseries databases) and assisting their reliability.

That said, I am open to any interesting problems in the broader Systems area.

Contact me through the email address in the menu.