1 day ago
Zürich, SwitzerlandMid Level
Responsibilities
- Turn research prototypes into tested, reusable data-generation, training, and evaluation pipelines.
- Build distributed experiment infrastructure for data loading, checkpointing, and collecting agent interactions.
- Profile workloads and improve GPU utilization, memory efficiency, and data throughput.
- Develop tests, experiment-tracking, and debugging tools while preserving research results.
- Collaborate with researchers to translate ideas into robust, maintainable software.
Requirements
- Strong Python skills and hands-on experience with PyTorch, JAX, or a comparable machine-learning framework.
- Experience building and debugging software for model training, inference, or large-scale data processing.
- Understanding of machine-learning experiments and how data, numerical precision, and implementation choices affect results.
- Sound software-engineering practices, including testing, profiling, version control, and documentation.
- Distributed training or data-processing experience with PyTorch distributed, DeepSpeed, or Ray is valuable.
- Familiarity with model-serving, machine-learning, or experiment-tracking tools is a plus; reinforcement-learning systems and multimodal datasets are also useful.
Benefits
- Equity is provided.
- Visa sponsorship is available.
- On-site work in Zurich, Switzerland, or San Francisco, California.