HPC Optimization Engineer

As a Software Engineer and play a key role in advancing scientific discovery through high-performance and accelerated computing. In this position, you'll contribute to a variety of cutting-edge open-source projects, working closely with researchers and developers across multiple scientific domains. Your work will focus on leveraging hardware accelerators-especially GPUs-to enhance performance and portability across platforms.

What You'll Do:

As part of a collaborative and forward-thinking team, your day-to-day will involve:

Deploying and fine-tuning machine learning tools and frameworks to run efficiently on modern hardware

Porting HPC applications and libraries to GPUs and optimizing them for peak performance

Researching and sharing best practices for unified programming models that enable portable, accelerator-agnostic code

Preparing scientific codes for deployment on state-of-the-art supercomputers with GPU-centric architectures

Engaging in robust software engineering practices such as version control (Git), automated testing, continuous integration, documentation, and debugging

You'll report to the Deputy Director of the Scientific Computing Core and be immersed in an environment that values innovation, collaboration, and scientific impact.

Who You Are:

We're looking for a detail-oriented and technically curious engineer who enjoys solving complex problems and working on impactful projects.

Required Qualifications:

A degree in Computer Science or a related technical field (B.S., M.S., or Ph.D.)

3+ years of hands-on software development experience

Strong proficiency in C, C++, and Python

Familiarity with additional languages like Julia, MATLAB, or R is a plus

Experience with parallel computing frameworks such as MPI and OpenMP

Practical GPU programming skills (e.g., CUDA or similar)

Knowledge of performance tuning tools (e.g., profilers, monitors, nvidia-smi, perf, etc.)

Experience with scientific libraries like BLAS, LAPACK, and FFT

Familiarity with portable programming models (OpenCL, SYCL, OpenMP offloading)

Comfortable working in Linux-based HPC environments; experience with SLURM is advantageous

Experience handling large-scale data (terabyte+ datasets)

Additional Skills That Make You Stand Out:

Solid understanding of software engineering principles, including SDLC, code reviews, and version control tools (e.g., Git, Mercurial)

Enthusiastic about learning new technologies, languages, and computing environments

Curious, collaborative, and customer-focused

Strong communication skills, both verbal and written