微信搜索-LLM大模型推理研发高级工程师-研发优化
Core
Develop and optimize online inference models (SLM, LLM) for WeChat search scenarios to ensure engine performance, stability, and cost-efficiency under high C-end traffic.
Role type
Senior IC LLM inference engineering specialist
Builds
High-performance inference engines for generative and discriminative large models
Domain
Search & Recommendation (Search, Ads, Recommendation)
Deliverable
production ML models
Required skills
C/C++, Linux development, distributed systems principles, LLM algorithm architecture, VLLM, SGLang
Preferred skills
Inference engineering optimization in search/ads/recommendation scenarios
Technologies
VLLM, SGLang
Responsibilities
Develop and optimize online inference models for various sizes; Apply distributed system principles to solve engineering challenges in LLM inference; Implement cutting-edge LLM Infra technologies; Collaborate with search algorithms to design generational model improvements
Seniority
Senior, hands-on IC
