Scientific Developer for Model Optimization
Lynker Corporation22 hours ago
College Park, MD, USASenior
Responsibilities
- Identify and resolve performance bottlenecks in the UFS Weather Model, MPAS atmospheric dynamic core, and coupled components.
- Optimize model performance through code optimization, build configuration, heterogeneous computing, and portable CPU/GPU programming.
- Implement runtime profiling for large parallel applications, covering CPU and GPU performance, memory, cache usage, and disk I/O.
- Develop configurable and reproducible profiling and optimization frameworks and workflows.
- Integrate optimized, GPU-enabled, and profiling-enabled components into the UFS build system.
- Port and test accelerated code on appropriate computing systems while ensuring portability and reproducibility.
- Investigate bit-for-bit reproducibility across CPU and GPU optimizations and implement reproducibility tests.
- Document profiling, instrumentation, optimization, and GPU or portable programming practices.
- Collaborate with teams supporting models and forecast workflows by delivering optimized code and profiling frameworks.
- Provide technical guidance and mentoring on profiling, optimization, and GPU or portable code development.
- Explore and validate the use of GenAI and LLMs to improve code quality and software development efficiency.
Requirements
- Bachelor's or master's degree in meteorology, atmospheric science, physical sciences, mathematics, computer science, software engineering, scientific computing, or a closely related field, with demonstrated numerical modeling or related scientific computing experience.
- Demonstrated experience with high-performance computing, software optimization, and building, running, debugging, and profiling large-scale parallel applications.
- Proficiency in C, C++, and/or Fortran and MPI.
- Familiarity with GPU or heterogeneous programming and portability frameworks such as CUDA, OpenACC, OpenMP target offloading, Kokkos, RAJA, PyTorch, or JAX.
- Familiarity with profiling concepts or tools such as gprof, TAU, or Scalasca.
- Experience with Linux/Unix shell scripting and automation.
- Experience with build systems such as CMake, Meson, autotools, or make is strongly preferred; strong configurable scripting and automation proficiency is acceptable.
- Proficiency with Git or similar code collaboration and management tools.
- Demonstrated ability to apply software engineering concepts and best practices.
- Preferred qualifications include a Ph.D. in a quantitative field, research computing experience, UFS or MPAS experience, atmospheric modeling experience, NWP workflow familiarity, HPC workflow and package management experience, GenAI/LLM workflow experience, and familiarity with continuous integration workflows and frameworks.
Benefits
- Hybrid schedule with two days per week in the College Park, Maryland office and three days working from home.
- Comprehensive healthcare, including medical, prescription drug, dental, and vision coverage, with no monthly employee cost.
- Paid time off and paid holidays.
- 401(k) retirement plan with company matching and an Employee Stock Ownership Plan.
- Flexible spending accounts, employee assistance program, short- and long-term disability insurance, and life and accident insurance.
- Tuition assistance, training and workforce improvement reimbursement, learning management resources, and personalized career growth plans.
- Spot bonuses, annual employee recognition awards, and an employee referral program.
- Team-oriented work environment at an employee-owned company.
About Lynker Corporation
Lynker Corporation provides scientific and technical consulting, geospatial solutions, and software development to U.S. environmental and water-resource agencies under services contracts. The company supports NOAA and the National Weather Service with meteorology, GIS, and forecasting tool development, and delivers program and portfolio management across conservation and resource programs. Founded in 2007 and headquartered in Leesburg, Virginia, Lynker is a privately held contractor.