Sort by
Refine Your Search
-
Category
-
Program
-
Employer
-
Field
-
deployed through conventional casting for standard applications and through 3D printing for complex or non-planar geometries. In parallel, circular end-of-life strategies will recover fibres, upcycle fines
-
structural interventions. It can be deployed through conventional casting for standard applications and through 3D printing for complex or non-planar geometries. In parallel, circular end-of-life strategies
-
concepts such as data parallelism, tensor parallelism, pipeline parallelism, checkpointing, and GPU communication. Experience with frameworks such as veRL, slime, Megatron-LM, DeepSpeed, TRL, vLLM, SGLang
-
, ideally in a technical or CS-related field Strong organizational skills and comfort managing multiple parallel timelines and stakeholders Working familiarity with tools such as Moodle, GitHub/GitHub
-
several projects moving in parallel. You communicate clearly in English, in writing and in negotiation. German is a plus, not a requirement. Workplace Workplace We offer A role with genuine ownership. You
-
This means experience with frameworks such as veRL, slime, Megatron-LM, DeepSpeed, TRL, vLLM, SGLang, or similar tools Strongly preferred Familiarity with distributed training concepts such as data parallelism
-
of vector query processing, serverless/just-in-time query execution, parallel and distributed data management, and data management systems on emerging hardware Passionate about inventing, building, and
-
and analysis methods for realistic CO2 electrocatalysis, with a focus on parallel investigations and accelerated aging. Dissect degradation processes of CO2 electroreduction catalyst, electrodes, and
-
between the moment data lands on disk and the moment a researcher — or a training job — can actually trust it, find it, and use it. Our parallel filesystems and object stores already move data fast. What
-
for Apertus models, from checkpoints during training to released models Make evaluations run quickly and at scale: parallel execution on Alps, efficient use of inference backends, caching, and result tracking