Skip to main content

CV

Will Paik
Author
Will Paik
I optimize large-scale GPU clusters for AI/ML workloads. Outside of work, I build a mini-supercomputer from consumer hardware and document every step of it here.

HPC Engineer with 9+ years of experience in system architecture and performance optimization for scientific and AI/ML workloads. Focus on translating hardware capabilities into real-world application performance, particularly for large-scale AI/ML training. Experienced in supporting diverse research needs across university clusters, statewide research infrastructure, and custom-built systems.


Experience
#

HPC Performance Engineer Research Computing, Northeastern University | Jan 2025 – Present

  • Optimizing multi-node PyTorch DDP training on production GPU clusters, tuning NCCL configuration, data loading, and Slurm resource binding
  • Building internal tooling that validates cluster health and surfaces misconfigured or degraded nodes
  • Supporting researchers with computational workflow optimization across multiple disciplines

GPU Training Benchmarking, AICR Benchmarking Group Massachusetts AI Computing Resource (AICR) | 2026

  • Contributed to a statewide AI research infrastructure initiative run separately from Northeastern’s own clusters
  • Built an automated multi-node GPU validation suite (PyTorch DDP, NCCL) that confirms nodes operate within expected ranges and surfaces misconfigured or degraded hardware
  • Target systems include B200 and RTX PRO 6000 nodes across multi-node configurations

HPC Software Consultant Institute for Computational and Data Sciences, Penn State University | Jan 2017 – Dec 2024

  • 8 years supporting 500+ researchers across multiple disciplines
  • Cluster performance optimization and user support
  • Developed containerized environments for reproducible research (Singularity Hub contributor)
  • System monitoring, resource allocation, and job optimization

Parallel Computing Support Application Engineer (Internship) MathWorks | Summer 2021

  • Optimized parallel computing toolboxes for MATLAB
  • Developed performance benchmarks for distributed computing
  • Created documentation for HPC integration with MATLAB

Technical Skills
#

Category Tools
Schedulers Slurm, PBS (job arrays, dependency chains, resource optimization)
Parallel Computing MPI (OpenMPI, Intel MPI), OpenMP, CUDA
Storage NFS (exports, home/scratch), storage layout and data management strategies
Containerization Singularity/Apptainer, Docker, Podman
Automation Ansible, Bash scripting, system provisioning
Monitoring & Validation Performance metrics, profiling, automated cluster health checking (HPL, PyTorch DDP validation)
Languages Python, C/C++, MATLAB, Shell scripting; Fortran (reading and debugging legacy scientific code)
Version Control Git
Performance Profiling, optimization, bottleneck analysis

Projects
#

Project Description
HPC From Scratch 6-node cluster from consumer hardware. Slurm, Ansible, NFS, FreeIPA, Lmod.
PyTorch DDP Benchmark Multi-GPU/multi-node distributed training scaling benchmark for HPC clusters. [GitHub]
pkg_audit RPM package consistency audit tool with Slurm partition sweep and Ansible remediation. [GitHub]
4D LiDAR SLAM Optimization Parallelized point cloud processing for real-time ROS 2 performance
Side Projects (upcoming) Game of Life web app, browser poker, ESP32-P4 thermal camera

Other Engineering Projects

NIST First Responder UAS Indoor Challenge (2022) Award: 3rd place + First Responder’s Choice (Prize: $80,000) Custom quadcopter for GPS-denied indoor emergency scenarios. [nist.gov]

VFS Design-Build-Vertical Flight Student Competition (2021 and 2022) Award: 3rd place (2022), 1st place in preliminary reports + Best Computational Simulation Award (2021). [engr.psu.edu]

9th and 10th ESA Global Trajectory Optimization Competition (2017 and 2019) Developed parallel algorithms for complex trajectory optimization. [psu.edu]


Education
#

The Pennsylvania State University

Degree Year Notes
PhD, Aerospace Engineering 2024 Minor in Computational Science. Dissertation: Multiple Gravity-Assist Trajectory Design with Continuous-Thrust Synergetic Maneuvers
MS, Aerospace Engineering 2015 Minor in Computational Science. Thesis: Optimal Orbit Raising via Particle Swarm Optimization
BS, Aerospace Engineering 2013

Talks & Workshops
#

See the Talks page for details.

Talk Venue Year
Introduction to Software Installation Northeastern University Summer 2026
Introduction to Parallel Computing Northeastern University Spring 2026
Linux Essentials for HPC Researchers Northeastern University Spring 2026
Teaching Assistant – Aerospace Analysis, Programming for Engineers Penn State University 2015–2016