Skip to main content
T0 Big tech Tencent New grad Rounds 1-2

Tencent AI Infra campus hiring first and second rounds (1)

Real Tencent AI Infra campus hiring first- and second-round interview questions covering operator optimization and related topics

Operator optimization

First round

Inference optimization

  1. Compare source-level differences among TensorRT-LLM, AGI, and vLLM
  2. What is the design motivation for continuous batching
  3. Introduce the source structure and implementation approach of fastllm.cpp

Fundamentals

  1. For CPU-bound work in Python, should you use multiprocessing or multithreading
  2. How inheritance is implemented under the hood in C++
  3. Common ways to speed up YOLO model inference

Coding

  1. Maximum subarray sum

Second round

Inference optimization

  1. Introduce the core ideas of vLLM and PagedAttention
  2. Approach to implementing PagedAttention with Triton

Fundamentals

  1. Hierarchy and traits of the CUDA memory model
  2. How to run work at compile time in C++ (constexpr and related)

Coding

  1. Count the number of 1 bits in an integer’s binary representation