Hybrid HPC Scientist - Model Optimization & GPU Profiling

Lynker Corporation; College Park (MD); Hybrid; USD 110,000 - 170,000; Full time

Benefits offered by this job

Comprehensive healthcare for employee

PTO and paid holidays

401(k) with company match

ESOP - employee ownership

Tuition assistance

Spot bonuses

Annual recognition awards

Employee referral program

Learning Management System access

Career growth plans

Job summary

Lynker Corporation in the DC Metro area is seeking a Scientific Developer for Model Optimization to advance performance of the UFS MPAS‑based workflows. The role offers a hybrid schedule with two days in our College Park, MD office and three days remote.

You will identify bottlenecks, implement profiling, and develop portable, GPU‑accelerated solutions while mentoring colleagues and collaborating with cross‑functional teams. A strong HPC/software background is required.

Qualifications

• Bachelor's or Master's in meteorology, atmospheric science, physical sciences, mathematics, computer science, software engineering, scientific computing, or closely related field with modeling experience.
• Experience in HPC, software optimization, and profiling large‑scale parallel applications.
• Proficiency in C, C++, and/or Fortran, as well as MPI.
• Familiarity with GPU/heterogeneous programming and portability frameworks (CUDA, OpenACC, OpenMP, Kokkos, RAJA, PyTorch, JAX).
• Familiarity with profiling concepts and tools (gprof, TAU, Scalasca).
• Experience with Linux/Unix shell scripting and automation.
• Experience with build systems (CMake, Meson, autotools, make).
• Demonstrated proficiency with Git for code collaboration.

Responsibilities

• Identify and resolve performance bottlenecks in the UFS MPAS workflow.
• Implement runtime profiling for CPU/GPU, memory, cache, and I/O.
• Develop configurable workflows for performance testing.
• Integrate optimized components into the UFS build system and ensure portability.
• Test and port code on appropriate HPC systems with reproducibility.
• Collaborate with teams to deliver profiling/instrumentation tools and guidance.
• Explore GenAI/LLMs to improve code quality and development efficiency.

Skills

High-performance computing

C/C++/Fortran

MPI

GPU programming

Linux scripting

Git version control

Software engineering

GenAI/LLM familiarity

Education

Bachelor's or Master's in relevant field

PhD in quantitative field (ideal)

Tools

CMake

Meson

Autotools

Profiling tools (gprof/TAU/Scalasca)

ecFlow/Rocoto/Cylc

Spack/pip

Git

Job description

Lynker Corporation in the DC Metro area is seeking a Scientific Developer for Model Optimization to advance performance of the UFS MPAS‑based workflows. The role offers a hybrid schedule with two days in our College Park, MD office and three days remote.

You will identify bottlenecks, implement profiling, and develop portable, GPU‑accelerated solutions while mentoring colleagues and collaborating with cross‑functional teams. A strong HPC/software background is required.