AI性能优化专家-计算
Core
Analyze AI infrastructure performance, build performance benchmarks and competitiveness models, and optimize underlying systems for training and inference in cloud-based AIGC scenarios.
Role type
Senior IC AI infrastructure performance engineer
Builds
Optimized AI training and inference workloads on cloud infrastructure using domestic accelerators
Domain
Cloud computing, AI infrastructure, heterogeneous computing
Deliverable
production ML models
Required skills
Linux system programming, C/C++, Python, Go, distributed system development, LLM/CV/NLP/Recommendation system principles, deep learning framework internals (TensorFlow/PyTorch), heterogeneous computing performance analysis, hardware-software co-optimization
Preferred skills
Cloud vendor AI infrastructure solutions, LLM training/inference optimization, model quantization and accuracy tuning, TensorRT/CUTLASS/Triton development, domestic AI accelerator ecosystem experience
