HPC systems administration
Operate Linux/Slurm research computing systems, monitor queues and node health, manage access, troubleshoot user issues, and keep shared cluster environments usable for research groups.
Technical Profile
Research computing, scientific software, and HPC operations
I build and maintain Linux computing environments, software stacks, and automation that make research workflows easier to run, debug, and reproduce.
At CSU Northridge, I managed two NSF-funded HPC clusters for the College of Science and Mathematics, supporting Slurm scheduling, Linux administration, software environments, GPU-capable nodes, and day-to-day user troubleshooting.
Linux
Slurm
Python
PyTorch
Bash
CUDA environments
Git
User support
Research Computing Stack
Operations to analysis
Access
Users, groups, SSH, permissions
Scheduler
Slurm queues, jobs, node state
Software
Compilers, Python, CUDA, research codes
Workflows
Scripts, logs, parsing, reproducible runs
Core Skills
Scientific software environments
Install, configure, and debug software stacks for computational chemistry, Python analysis, GPU-enabled workloads, and shared research workflows.
Automation and scripting
Write shell and Python tools for job setup, filesystem organization, log inspection, data parsing, and repeatable command-line workflows.
Machine learning workflows
Build and run PyTorch training and inference workflows, manage checkpoints and logs, prepare tensor data, and analyze model outputs on GPU-backed systems.
Representative Technical Work
Research computing operations
CSUN HPC cluster administration
- Scope
- Managed two NSF-funded HPC clusters for the CSUN College of Science and Mathematics.
- Work
- Supported Slurm scheduling, Linux access, shared software environments, node checks, and user troubleshooting.
- Tools
- Linux, Slurm, Bash, SSH, CUDA-aware environments, Git, remote administration.
Software support
Shared scientific computing environments
- Scope
- Kept research software usable across shared machines and changing project needs.
- Work
- Resolved dependency issues, configured Python environments, supported compiled codes, and documented working setups.
- Tools
- Python, Conda, compilers, CUDA, VASP, Quantum ESPRESSO, pymatgen, ASE.
Automation
Command-line workflow tooling
- Scope
- Built practical scripts around recurring research-computing tasks.
- Work
- Automated file setup, queue checks, log summaries, output validation, and data cleanup.
- Tools
- Bash, Python, Slurm commands, regular expressions, structured data parsing.
Tools Matrix
Systems administration
Linux, Slurm, SSH, users/groups, permissions, storage, queue monitoring, remote administration
GPU and software environments
NVIDIA/CUDA environments, compilers, Conda, Python packaging, dependency debugging
Scientific programming
Python, PyTorch, NumPy, Pandas, Matplotlib, Jupyter, Bash, command-line tooling
Chemistry software
VASP, Quantum ESPRESSO, pymatgen, ASE, VESTA, Sumo, structure and output parsing
Documentation and collaboration
Git, GitHub, Markdown, HTML/CSS, reproducible notes, user-facing setup instructions
Technical work that keeps research moving
I treat computing infrastructure, software environments, automation, and documentation as part of the research workflow: the system should be reliable enough that the science can stay in focus.