About Me
Here is Tianyu Feng (Tim, 冯天昱).
I am a PhD student in Computer Science and Engineering at HKUST, entering my second year in Sep 2026, advised by Prof. Wei WANG.
Formerly I got my master’s in Computer Science and Technology in HiPO Group at Beihang University, advised by Prof. Hailong Yang.
I would always be happy to collaborate! Email me at - tian[dot]yu[dot]feng[at]outlook[dot]com
Research Background
- Sep 2025 - now: PhD @ HKUST CSE (The Hong Kong Univ. of Sci. & Tech.)
- Apr 2025 - now: Research Intern @ Alibaba Group (Aicheng Technology, TRE)
- Feb 2025 - Aug 2025: Research Assistant @ HKUST CSE
- Sep 2022 - June 2025: MEng @ Beihang University (BUAA) (CS) (计算机科学与技术, 学术型硕士)
- Sep 2018 - June 2022: BEng @ Beihang University (BUAA) (CS) (计算机科学与技术, 工学学士)
Research Interests
- Distributed System Optimization: scheduling, caching and data placement in large-scale distributed systems.
- Parallel Computing: automatic kernel generation and optimization on GPUs.
Selected Publications
A full list is on my publications page. * denotes equal contribution.
- Format-Driven Automatic Pipeline Construction and Load Balancing for SpMM on GPUs
Kaige Zhang*, Tianyu Feng*, Xin You, et al. — SC 2026, Best Paper / Best Student Paper Finalist - DiFlow: A System for Micro-Serving Text-to-Image Diffusion Workflows
Lingyun Yang*, Suyi Li*, Tianyu Feng, et al. — SOSP 2026 - FlashPS: Efficient Generative Image Editing with Mask-aware Caching and Scheduling
Xiaoxiao Jiang*, Suyi Li*, Lingyun Yang, Tianyu Feng, et al. — EuroSys 2026 - AtRec: Accelerating Recommendation Model Training on CPUs
Siqi Wang*, Tianyu Feng*, Hailong Yang, et al. — TPDS 2024 - dgQuEST: Accelerating Large Scale Quantum Circuit Simulation through Hybrid CPU-GPU Memory Hierarchies
Tianyu Feng, Siyan Chen, Xin You, et al. — NPC 2021, Best Paper
News and Updates
- Jul 2026: Sparsene is nominated as a finalist for both the Best Paper Award and the Best Student Paper Award at SC 2026, in the Programming Frameworks area. See you in Chicago!
- Jul 2026: One work (DiFlow) is accepted by SOSP 2026.
- Jul 2026: One work (Sparsene) is accepted by SC 2026. Great thanks to my advisor Prof. Hailong Yang and my co-first author Kaige Zhang!
- Feb 2026: One work (Drawloom) was presented at PPoPP 2026 in Sydney.
- Aug 2025: One work (FlashPS) is accepted by EuroSys 2026.
- Jun 2025: I have successfully defended my master’s thesis and earned the MEng degree. Sincere appreciation is extended to my advisor Prof. Hailong Yang for his guidance and support throughout this academic journey.
- Apr 2025: Offered HKPF (Hong Kong PhD Fellowship) 25/26. Sincere gratitude to all who supported the application process.
- March 2024: One work (AtRec [source code]) is accepted by IEEE Transactions on Parallel and Distributed Systems (TPDS). Great thanks to my advisor Prof. Hailong Yang and my excellent collaborator Siqi Wang!
- Sep 2023: Awarded First Prize of Beihang School of Computer Science and Engineering Academics Scholarship.
- Aug 2023: Our team won Third Prize in SolverChallenge23. Thanks to my advisor Prof. Hailong Yang and my teammate Kelun Lei.
- Dec 2022: Our team won Second Prize (rank 2/45) in Tianchi Competition: DeepRec CTR Model Performance Optimization. Thanks to my teammate Siqi Wang!
- Oct 2022: Awarded National Scholarship.
- Sep 2022: Awarded First Prize of Beihang School of Computer Science and Engineering Academics Scholarship.
- Nov 2021: dgQuEST was selected as the best paper in NPC 2021!
- Sep 2021: One work (dgQuEST [source code]) was accepted by NPC 2021. Great thanks to my advisor Prof. Hailong Yang and Dr. Xin You!
- May 2021: Our team won First Prize and Application Innovation Award (highest score in quantum circuit simulation) in ASC20-21. Great thanks to my teammates Siyan Chen, Quqi Liu, Jiwei Hao, Chun Cai and our advisor Prof. Hailong Yang!
Programming Languages, Frameworks and Other Skills
- Programming/Script Languages: C/C++, Python, Bash
- Parallel Programming Models: CUDA, CUTLASS (CuTe), MPI, OpenMP, AVX / ARM NEON / RISC-V RVV
- Profiling and Analysis: Nsight Systems, Nsight Compute, perf
- Utilities/Systems/Tools: Git, Docker, Linux, CMake, Spack
- AI/ML Frameworks: PyTorch, TensorFlow