cv
Education, experience, publications, projects, and skills. A PDF copy is available from the icon at the top right.
Basics
| Name | Benjamin Cobb |
| Label | Incoming HPC Engineer, Citadel Securities |
| benjamincobb00@gmail.com | |
| Url | https://www.ben-cobb.com |
| Summary | PhD in Computer Science from the Georgia Institute of Technology (August 2026), with research in numerical linear algebra, tensor decompositions, and high-performance computing. Joining Citadel Securities as an HPC Engineer in November 2026. |
Work
-
2026.11 - Present -
2024.06 - 2025.05 Discrete Algorithms Intern
Oak Ridge National Laboratory
Worked on discrete algorithms during internships at ORNL.
-
2020.06 - 2020.08 Computer Science Research Intern
Sandia National Laboratories
Internship during the pandemic, conducted virtually.
-
2019.01 - 2026.08 Graduate Research Assistant
Georgia Institute of Technology
Researched numerical linear algebra, tensor decompositions, and high-performance computing.
-
2016.01 - 2019.01
Education
Publications
-
2025.12.01 -
2024.01.01 On Rank Selection for Nonnegative Matrix Factorization
IEEE Big Data
-
2024.01.01 -
2023.06.01 Distributed-Memory Parallel JointNMF
International Conference on Supercomputing, ICS 2023
Skills
| High-Performance Computing | |
| MPI | |
| CUDA | |
| Kokkos | |
| OpenMP | |
| cache aware programming |
| Programming | |
| C/C++ | |
| MATLAB | |
| Python |
| Machine Learning | |
| Deep Learning | |
| Neural Nets | |
| PyTorch | |
| Matrix and Tensor Factorizations |
| Software Development | |
| Git | |
| CMake | |
| Unit Tests | |
| Vim | |
| Slurm | |
| Tmux | |
| Linux Command Line | |
| Scripting |
| Performance Analysis and Debugging | |
| Roofline Bandwidth Analysis | |
| VTune | |
| GDB |
Projects
- - Present
PLANC
Distributed MPI software package for large-scale matrix and tensor factorizations used in data analysis and compression.
- - Present
GenTen Portable Tensor Decompositions
Performance portable software package for tensor factorizations used in data compression.
- - Present
MATLAB Tensor Toolbox
Implemented minimum cost bipartite matching algorithm for least-squares cosine differences.