Sort by
Refine Your Search
-
Employer
- Argonne
- Oak Ridge National Laboratory
- Duke University
- Harvard University
- Rutgers University
- SUNY Polytechnic Institute
- Stanford University
- University of Miami
- University of Nebraska Medical Center
- University of North Carolina at Chapel Hill
- University of Utah
- Brookhaven National Laboratory
- Northeastern University
- Texas A&M University
- University of Florida
- University of New Hampshire
- 6 more »
- « less
-
Field
-
adaptive optimization during needle insertion, integrating live ultrasound imaging with GPU-accelerated dose calculation and optimization. The Postdoctoral Research Associate will join a multidisciplinary
-
computing environments, and GPU programming. Necessary skills include knowledge of data processing using software (e.g., Matlab, R, IDL) and/or statistical/mathematical programming languages (e.g., R, Matlab
-
and GPU-accelerated tools for circuit and system design optimization, addressing challenges in physical design, timing analysis, and large-scale hardware design automation. The researcher will
-
simulation methods, GPU-accelerated computations, several programming languages, and presenting results to wide technical and non-technical audiences. Additionally, the candidate will also develop theory and
-
). Practical experience with cloud computing platforms (e.g., AWS, GCP, Azure). Additional Qualifications Experience with multi-GPU model training and large-scale inference. Familiarity with modern AI
-
in GPU programming one or more parallel computing models, including SYCL, CUDA, HIP, or OpenMP Experience with scientific computing and software development on HPC systems Ability to conduct
-
with OFDM modulation required. Skills Programming skills in MATLAB and or Python required, experience with wireless testbeds desirable, some familiarity with GPU programming desirable (to support
-
/TimeSformer, CLIP/BLIP or similar) in PyTorch, including scalable training on GPUs and reproducible experimentation. Demonstrated experience building explainable models (e.g., concept bottlenecks, prototype
-
scientists and engineers are accustomed to. Moreover, the vast majority of the performance associated with these reduced precision formats resides on special hardware units such as tensor cores on NVIDIA GPUs
-
disease insights. The lab has state-of-the-art computing capabilities with an in-house cluster serving 80 CPU cores and 1.5TB of RAM, as well as a newly acquired NVIDIA DGX box with eight H100 GPUs and 224