Skip to main content
T0 Big tech Tencent Internship Round 1

Tencent AI Infra internship first round (1)

Real Tencent AI Infra internship first-round interview questions covering inference optimization, operator optimization, high-performance computing, and related topics

Inference optimization Operator optimization High-performance computing

Project experience

  1. Introduce project experience and related technical details

Inference optimization

  1. Explain the design motivation of chunked prefill and the core problem it solves
  2. In a large cluster (NVLink within a node, RDMA between some machines across nodes), how would you design a distributed inference plan

High-performance computing

  1. Compare reduce-scatter and all-to-all collective communication
  2. What methods can reduce launch kernel overhead
  3. Causes of bank conflict in CUDA programming and how to fix them

Coding

  1. Reverse nodes in k-Group