Education
PhD in Computer Science (In Progress)
Tennessee Technological University, Cookeville, TN
Dissertation: Deep Reinforcement Learning for NCCL Optimization
Research Experience
Graduate Research Assistant
Tennessee Technological University
DynamICCL Project
- Developing deep reinforcement learning approaches for optimizing NCCL collective operations
- Conducting large-scale experiments on cloud infrastructure (Chameleon Cloud, GCP)
- Performance analysis of distributed GPU training under network congestion
- Implementing custom NCCL tuner plugins using C++/Python
Technical Skills
Programming Languages:
Python, C++, Bash, CUDA
Frameworks & Libraries:
PyTorch, NCCL, OpenMPI, NumPy, Pandas
Tools & Platforms:
Git, Docker, Linux, Google Cloud Platform, Chameleon Cloud
Specializations:
Distributed Computing, GPU Programming, Deep Learning, Network Analysis,
High-Performance Computing
Projects
DynamICCL
Deep reinforcement learning optimization of NVIDIA NCCL for distributed training.
NCCL Benchmarking Suite
Custom Python framework for measuring AllReduce performance across different configurations and network conditions.
Contact
For a complete CV with references and detailed project descriptions, please contact me.
Download: CV (PDF) (coming soon)