About me

I’m a PhD student in Computer Science at Florida State University, where my research focuses on high-speed interconnects for HPC and AI/ML workloads. My work explores how software-defined networking (SDN) and local router intelligence can improve routing efficiency and mitigate congestion in large-scale systems. My research interests includes Network Interconnect, Adaptive Routing, Communication Optimization and MPI.
I am honored to be advised by Professor Xin Yuan. I completed my undergraduate from Tribhuvan University, Nepal where my undergraduate research advisor was Professor Subarna Shakya. You can see my publications in the publication
tab or at google scholar for my most recent publications.
Key highlights of my research include:
✅ Developing SDN-based adaptive routing strategies for Dragonfly topologies
✅ Enhancing UGAL routing with accurate local latency estimation
✅ Applying deep learning for congestion-aware routing, matching and outperforming traditional adaptive schemes
I’ve contributed to simulators like Booksim, CODES, and SST, and my work has been published at venues like SC, ICS, and CCGRID.
🛠️ Tools & Tech: Python, C/C++, MPI, CUDA, OpenMP, PyTorch, TensorFlow, SDN, SST, CODES, Booksim I’m excited about pushing the limits of interconnect design for the next generation of AI and HPC systems.
I enjoy Soccer, Basketball, Chess, and Traveling during my free time.
Recent News!!!
| Honored to receive the Best Research Presentation Award at the CS Expo organized by the Florida State University Department of Computer Science. |
| Paper Accepted in CCGRID-25: Exploiting Software-Defined Networking Technology for Improving UGAL Routing in Dragonfly Networks |
| Poster Accepted in SC-24: Exploring Software-Defined Networking for Routing in Dragonfly Networks |
| Paper published in ICS-24: Enhanced UGAL routing schemes for dragonfly networks |
| Honored to receive the Outstanding Research Assistant Award for 2023-2024 from the Computer Science Department, FSU |
| Poster Accepted in SC-20: Achieving the performance of global adaptive routing using local information on dragonfly through deep learning |