HPC Scientific Software Engineer - Research Computing

About

We, at IT@JH Research Computing, are looking for a HPC Sr. Scientific Software Engineer to join our team and contribute to the design, construction, and support of Johns Hopkins Universitys high-performance computing and AI research infrastructure. We provide a dynamic environment that brings together systems and software engineering to deliver scalable and reproducible solutions for data-intensive research. Our team thrives on collaboration, continuous learning, and supporting innovative research initiatives. This full-time position offers a starting salary range of $99,800 to $175,000 annually, commensurate with experience, and is based at the Johns Hopkins Bayview campus, operating Monday to Friday from 8:30 AM to 5 PM.

Responsibilities

• Design and implement strategies for deploying scientific software on HPC and AI systems.
• Create computational workflows, selecting the most effective software configurations, utilizing tools such as Ansible for automation.
• Assist teams in tuning and optimizing AI models and gateway applications like XDMoD, Coldfront, Open OnDemand, CryoSPARC Live, SBGrid, and AI Agents.
• Analyze and enhance the performance of AI models and HPC applications, prioritizing GPU-enabled computing.
• Establish parallel processing, distributed computing, and resource management methods for efficient job execution.
• Develop, debug, and maintain software tools, libraries, and frameworks essential for HPC and AI tasks.
• Collaborate with system teams and software vendors including NVIDIA, Intel, and Matlab to optimize performance.
• Utilize CUDA, DNN, TensorRT, and Intel Compilers to boost system efficiency.
• Oversee scientific software deployment across HPC, cloud, and colocation facilities.
• Manage the installation, configuration, and upkeep of HPC packages using tools like CMake, Make, EasyBuild, Spack, and Lua module files.
• Engage closely with cross-functional teams, including researchers and software developers, to tackle complex HPC/AI problems.
• Mentor junior engineers and promote a culture of continuous learning.
• Resolve technical challenges and conduct root cause analyses for HPC/AI software issues.
• Implement solutions to enhance system reliability and prevent issues from reoccurring.
• Conduct training workshops for researchers and students on troubleshooting, workflow optimization, and utilizing HPC systems effectively.
• Remain updated on advancements in HPC and AI technologies and methodologies.
• Integrate new research into current systems to enhance performance and capabilities.
• Develop and oversee container orchestration strategies ensuring application scalability, reliability, and security.
• Create thorough documentation for system architectures, performance metrics, and project progress.
• Ensure adherence to security and regulatory requirements for all HPC and AI platforms.

Requirements

• PhD in a quantitative discipline.
• Five years of experience in HPC user support, software deployment, and performance optimization within an academic or research environment.
• Additional education may substitute for required experience and additional related experience may substitute for required education beyond a high school diploma as permitted by the JHU equivalency formula.
• Eight or more years of professional experience in high-performance computing, large-scale systems, or research software engineering (preferred).
• Deep proficiency in Linux systems administration, performance tuning, and automation tools such as Ansible, Terraform, Jenkins, or similar (preferred).
• Experience with cluster management, workload schedulers (e.g., Slurm), and distributed or parallel file systems (e.g., GPFS, Lustre, WekaFS, Ceph) (preferred).
• Strong programming or scripting skills in languages such as Python, Bash, C/C++, Go, or Rust (preferred).
• Familiarity with containerization and orchestration technologies used in HPC (e.g., Singularity, Apptainer, Docker, Kubernetes) (preferred).
• Understanding of high-speed interconnects (InfiniBand, 100/400 Gb Ethernet) and storage/data access patterns for AI and analytics (preferred).
• Experience in developing or maintaining CI/CD pipelines and module environments (Lmod/Spack) for research software (preferred).
• Knowledge of GPU computing (CUDA, ROCm), MPI/OpenMP, and AI/ML frameworks (preferred).
• Demonstrated ability to collaborate with researchers on performance optimization, workflow design, and reproducible computing (preferred).

Tech Stack

• AI
• Ansible
• Bash
• CI/CD
• Cloud
• Ceph
• CUDA
• Docker
• ELK
• Ethernet
• Grafana
• InfiniBand
• Support
• Jenkins
• Kubernetes
• Linux
• Matlab
• Prometheus
• Python
• Rust
• Security
• Terraform
• Machine-Learning

Location

Keswick Road 3910, Baltimore, United States

Salary

$99,800 - 175,000 per year

Benefits

• None explicitly listed.