-
spatial and temporal resolutions by analyzing point cloud and camera data You will develop a deep understanding of debris flow motion, and use this to perform hypothesis driven analysis of in-situ debris
-
Performance Computing (HPC) clusters and supercomputers. The project with limited project financing and therefore limited contract duration aims to bring cloud-like deployment, automation, and security
-
-class infrastructure is available, including a full range of experimental and clinical MR systems fully dedicated to research, state-of-the-art local and scalable cloud-based compute infrastructure (CPU
-
topic in the Chair of Engineering Geology at ETHZ As part of this, the group collects and analyses an unprecendented set of field datasets The collected data includes timelapse point clouds, video imagery
-
Lugano and in Zurich. For this position the work location is Lugano. Project background As high-performance computing, AI, and cloud technologies continue to converge; CSCS is redefining how advanced
-
delivery of the shared Geo-Intelligence Platform, including data access, processing pipelines, model-training environments, deployment capabilities and integration with HPC and cloud infrastructure
-
with LLM APIs, open models, evaluation pipelines, orchestration frameworks, and related tooling. Familiarity with model adaptation, fine-tuning, or related optimization methods. Experience with cloud
-
workflows using modern IaC tools Operate, troubleshoot, and improve Linux systems on both bare metal and virtual infrastructure Develop and maintain CI/CD pipelines Build and operate on-premise cloud services
-
services powering advanced scientific computing and AI/ML workloads at scale. You will work at the intersection of systems engineering, software engineering, cloud-native infrastructure, and high-performance
-
databases, cloud deployment (AWS/Azure/GCP), Docker) are also very welcome. You will need to be highly flexible and responsive to emerging opportunities with partners from industry. The candidate should be