📋 Job Overview

Tencent is seeking a Large Model Inference Optimization R&D Engineer specializing in operator optimization and compilation for locations in Shenzhen, Beijing, Shanghai, and Hangzhou. The role focuses on enhancing the performance of large model inference on GPU and AI chips through low-level optimizations and framework extensions. Responsibilities include developing high-performance operators, exploring cutting-edge technologies like MOE and dynamic computation graph optimizations, and improving computational efficiency in tools such as vLLM, SGLang, and PyTorch.