Physical Intelligence25.08.2024
Инженер инфраструктуры машинного обучения
Зарплата не указана
РЫНОК
16 333 ₽медиана по профессии
Data Scientist · 215 вакансий с указанной зарплатой
5 000половина предложений: 13 167–21 846120 000
Работодатель не указал зарплату — сравните с рынком сами.
Полная занятостьОфис
Обязанности
- 01Own training/inference infrastructure: Design, implement, and maintain systems for large-scale model training, including scheduling, job management, checkpointing, and metrics/logging
- 02Scale distributed training: Work with researchers to scale JAX-based training across TPU and GPU clusters with minimal friction
- 03Optimize performance: Profile and improve memory usage, device utilization, throughput, and distributed synchronization
- 04Enable rapid iteration: Build abstractions for launching, monitoring, debugging, and reproducing experiments
- 05Manage compute resources: Ensure efficient allocation and utilization of cloud-based GPU/TPU compute while controlling cost
- 06Partner with researchers: Translate research needs into infra capabilities and guide best practices for training at scale
- 07Contribute to core training code: Evolve JAX model and training code to support new architectures, modalities, and evaluation metrics
Требования
- 01Strong software engineering fundamentals and experience building ML training infrastructure or internal platforms
- 02Hands-on large-scale training experience in JAX (preferred), PyTorch
- 03Familiarity with distributed training, multi-host setups, data loaders, and evaluation pipelines
- 04Experience managing training workloads on cloud platforms (e.g., SLURM, Kubernetes, GCP TPU/GKE, AWS)
- 05Ability to debug and optimize performance bottlenecks across the training stack
- 06Strong cross-functional communication and ownership mindset