4th-yr Ph.D. @SysML-Princeton | Previously @google, @AWS, @MPI-INF, and @uw-mad-dash
specreason. PoC for "SpecReason: Fast and Accurate Inference-Time Compute via Speculative Reasoning" [NeurIPS '25]
73marconi. Artifact for "Marconi: Prefix Caching for the Era of Hybrid LLMs" [MLSys '25 Outstanding Paper Award, Honorable Mention]
63failfast. Artifact for "Fail Fast, Win Big: Rethinking the Drafting Strategy in Speculative Decoding via Diffusion LLMs" [arXiv '25]
20paper_notes. Personal blog + reading notes on system-ish papers
17torch_profiler. Simple PyTorch profiler that combines DeepSpeed Flops Profiler and TorchInfo
12cs759-sp21. CS/ECE/ME/EP 759 (High Performance Computing for Engineering Applications) Course Project: Cautiously Aggressive GPU Space Sharing to Improve Resource Utilization and Job Efficiency
8ruipeterpan.github.io. Rui's personal portfolio website
4cs639-fa20. CS 639 (Introduction to Computer Vision) Course Project: Real-time Vision Tasks with a High Degree of Freedom Robot Arm
2cs61c-su19. CS 61C: Great Ideas in Computer Architecture (Machine Structures) Course & Lab Projects
2echelonflow_slides.
2Fast_dLLM_v2_1.5B. Artifact for "Fail Fast, Win Big: Rethinking the Drafting Strategy in Speculative Decoding via Diffusion LLMs" [arXiv '25]
2cos598d_sp24. Python
1