aijobs.net

VLA训练infra算法工程师 - XiaomiRobotics

北京 R

CNY 240K-480K (estimate) Mid-level Full Time

Apply Save
Found 4h ago
Tasks
Perks/Benefits
Skills/Tech-stack

BF16 | C++ | CPU architecture | CUDA | Data-parallel | DeepSpeed | Distributed Data Parallel | Distributed Training | Distributed data | FP8 | FSDP | FlashAttention | GPU Architecture | Infiniband | Kernel optimization | Linux | Megatron-LM | Memory Management | Mixed Precision | Mixture of Experts | Model Parallelism | NCCL | Network fundamentals | Pipeline parallelism | PyTorch | Python | RDMA | Tensor Parallelism | Triton

Education

N/A

Roles

Deep Learning Engineer | Distributed Training Engineer | Engineer | Learning Engineer | Machine Learning Engineer | Training Engineer

Regions

Asia/Pacific

Countries

China

States

Beijing, CN

Cities

Beijing, Beijing, CN

Apply Save
Language: zh Views: 3 Clicks: 0 Saves: 0

Related jobs