Sort by
Refine Your Search
-
models, LLMs and Transformer architectures Excellent programming skills in PyTorch/JAX and experience working with GPUs and high-performance clusters. Strong mathematical skills with excellent
-
learning, multicore and GPU programming, and highly parallel systems. Good knowledge in one or more of the following programming languages/environments: C/C++, Python, PyTorch (or similar), and Cuda. Place
-
precision algorithms for CPUs and GPUs. Performance engineering and analysis including application profiling, benchmarking to identify performance bottlenecks. Verification, and validation of the developed
-
exposing hardware accelerators, such as GPUs and FPGAs, in a seamless and portable way. This includes designing execution logic and resource-scheduling strategies that make efficient use of available
-
tasks across distributed infrastructures. A key aspect of the position involves integrating and exposing hardware accelerators, such as GPUs and FPGAs, in a seamless and portable way. This includes
-
environment, which brings together more than 400 researchers across disciplines. The collaboration provides access to substantial computational resources (GPU nodes), advanced high-throughput instruments
-
together more than 400 researchers across disciplines. The collaboration provides access to substantial computational resources (GPU nodes), advanced high-throughput instruments (including a FACS, mass
-
access to substantial computational resources (GPU nodes), advanced high-throughput instruments (including a FACS, mass photometer, ITC, SPR, and others), and state-of-the-art characterization tools
-
management, high-performance computing systems, GPU acceleration, and parallel file systems * Documented experience with container and cloud technologies such as Docker, Helm, and Kubernetes * Ability
-
frameworks (e.g., PyTorch). Engineering skills: GPU/cluster training, experiment tracking, data engineering. Ability to formulate research questions, run empirical studies at scale. *for students with