Interview prep

Week 12 · Part of Kernel Ladder · Prev: capstone

What this page is for

Kernel interviews usually mix live coding, architecture questions, a performance-debugging case and a deep dive on your own work. Everything below comes from the course labs, so you are rehearsing work you have already done.


Live-coding drills

Forty-five minutes each, plain HIP, no autocomplete. Repeat two of them in CUDA if you are interviewing at CUDA-first teams.

Questions to answer in under two minutes

  1. Walk through an MI300X’s memory hierarchy — registers, LDS, L1, L2, Infinity Cache, HBM — and what its chiplet layout means for kernels.
  2. What limits occupancy on CDNA3, and when does higher occupancy hurt?
  3. Wave32 vs wave64: what changes in your reduction and shuffle code?
  4. MFMA vs WMMA: how do throughput, fragment layout and register use differ?
  5. Why is decode memory-bound and prefill compute-bound, and what does that mean for kernel design?
  6. Explain FlashAttention’s online softmax and why it saves HBM traffic.
  7. FP8 FNUZ vs OCP: what breaks if you mix them up? What does MXFP4 block scaling buy?
  8. A rocprof-compute report shows low matrix-core utilization and high memory-wait time. What do you try?
  9. A kernel is fast alone but slow inside the model. What do you check?
  10. How would you overlap an all-reduce with a GEMM?

Story bank

Prepare three stories, each as situation → action → numbers:

  • a hard performance bug you found and fixed
  • a time careful testing caught a numerics problem others missed
  • one result from this course: the GEMM ladder or the capstone

Mock interviews

Do at least two: one with a peer from GPU MODE or a local meetup, and one by handing your capstone write-up to a reviewer and asking to be grilled on every number.