Skip to main content
T1 Large companies and unicorns OPPO Internship Round 1

OPPO Cloud AI Infra internship first round

Real OPPO AI Infra internship first-round interview questions covering inference and operator optimization

Inference optimization Operator optimization

Project questions

  1. Briefly introduce related projects you worked on

Theory

  1. Facing a large-model inference framework you have never used, how do you get productive quickly?
  2. When converting data-type precision, how do you keep numerical stability?
  3. How are Warp/Block partitioning and the SIMT execution model related?

Open scenario

  1. As tech lead, you need to optimize a cloud CV model inference service: raise per-GPU throughput and cut latency. How would you approach it?