Sort by
Refine Your Search
-
Employer
- Forschungszentrum Jülich
- Fraunhofer-Gesellschaft
- Nature Careers
- Technical University of Munich
- Deutsches Elektronen-Synchrotron DESY
- Max Planck Institute for Intelligent Systems, Tübingen site, Tübingen
- Max Planck Institute of Animal Behavior, Radolfzell / Konstanz
- Max Planck Institute of Geoanthropology, Jena
- TECHNISCHE UNIVERSITAT DRESDEN (TU DRESDEN)
-
Field
-
Max Planck Institute for Intelligent Systems, Tübingen site, Tübingen | Bingen am Rhein, Rheinland Pfalz | Germany | 17 days ago
— operates a state-of-the-art GPU cluster with more than 1200 GPUs, serving as a critical backbone for advancing ground-breaking research in AI. Possible tasks include: Build, administer, optimize, and
-
, TrustLLM and EuroLingua-GPT, in which large foundation models are trained from scratch on the basis of several million GPU hours and several thousand GPUs. The distinctive feature of the work at the FMR-Lab
-
experiment operation in 2028. Emphasis has to be put on the application of the software in real-time, making use of massive parallelism on CPU and/or on GPU. Your profile: From the applicant, we expect a
-
researcher with a proven track record in areas relevant to auto-tuning, focusing on ML-driven compiler optimization, transfer learning, and programming for heterogeneous systems across CPUs, GPUs, and
-
researcher with a proven track record in areas relevant to auto-tuning, focusing on ML-driven compiler optimization, transfer learning, and programming for heterogeneous systems across CPUs, GPUs, and
-
, parallel/distributed computing, as well as diverse architectures and understanding of its impact on application performance Knowledge in GPU-based programming and modelling of scientific simulations
-
computers Desirable: Knowledge in standard HPC programming (C/C++, Fortran, MPI, OpenMP, GPU programming, etc.) Experience supporting application developers German language is an advantage You don’t need
-
, this position will primarily focus on optimising GPU utilisation for AI and molecular dynamics applications. About your role: Work closely with our industry partners on a high-performance computing (HPC) project
-
performance on neuromorphic hardware. What you will do Develop and train CNN and SNN models using frameworks like Keras, PyTorch, and SNNtorch Implement GPU acceleration using CUDA for efficient training
-
are an excellent team player and a curious researcher, knowledgeable within the field of HPC, for example with GPU-accelerated HPC systems or RISC-V processors. You know how to tackle complex