硬件加速训练AI Infra工程师-Data
Core
Develop self-researched hardware training frameworks and optimize distributed training for large models on proprietary hardware.
Role type
Senior IC hardware-accelerated AI infrastructure engineer (training)
Builds
Self-researched training frameworks, distributed training modes, and optimized operators for large models
Domain
AI Infrastructure / Large Language Models / Heterogeneous Computing
Deliverable
production ML models
Required skills
PyTorch, Megatron-LM, FSDP, distributed training (DP/SP/TP/PP), hardware accelerator training, operator development, compiler scheduling optimization
Preferred skills
Large-scale cluster training (100s-1000s of cards), NPU training/inference, computer architecture, GPU/FPGA/AI chip development
Technologies
Torch Compiler, Megatron, FSDP, DP, SP, TP, PP
Sourced via bytedance · Listed on CareerPlan, which tracks 855,000+ jobs from 20+ sources.
