AI Infra系统研发工程师-成都/上海
Core
Design and optimize large-scale distributed KV cache systems and inference frameworks for high-performance LLM serving.
Role type
Senior IC AI infrastructure engineer (LLM inference systems)
Builds
Distributed KV cache systems, optimized inference frameworks, and high-concurrency low-latency services
Domain
AI infrastructure, Large Language Models (LLM), High-Performance Computing (HPC)
Deliverable
production ML models
Required skills
C++, Python, System programming, Parallel computing, Memory management, Performance tuning, Transformer architecture, SGLang, vLLM, TensorRT-LLM, LightLLM, Docker, Kubernetes
Preferred skills
OS internals, GPU/CPU architecture, Interconnect buses
Responsibilities
Lead planning and architecture of distributed KV cache systems; Optimize H2D/D2H transmission performance across heterogeneous hardware; Optimize underlying communication libraries for KV cache and inter-card efficiency; Optimize cross-datacenter PD communication; Research and implement GPU confidential computing for inference security.
