AI Infra Interviews logo
← CUDA, Triton & Kernel Engineering / 18
hardNVIDIAFireworks

cuBLAS already gives you a fast GEMM. What does CUTLASS give you that it does not, and when is that worth it?

Use cuBLAS or cuBLASLt when its supported operation fits. CUTLASS exposes kernel building blocks for custom layouts, fusion and scheduling, but each choice needs a whole-operation benchmark and an ownership plan.

Updated Oct 2026 · Learn how AI infrastructure works through explanations, worked examples and diagrams, then practise applying it to interview questions.

Use cuBLAS or cuBLASLt when its supported operation fits. CUTLASS exposes kernel building blocks for custom layouts, fusion and scheduling, but each choice needs a whole-operation benchmark and an ownership plan.

This answer is included with a free account. Sign in to read it and save your progress.

free account · no card needed

Work through the numbers

Change the assumptions in a free calculator, then explain what the result establishes and what still needs measurement.

The concepts behind this question

Use these concept explanations to review the ideas behind this question.

Advanced
⚡ Kernels & Compilers🔒 Premium
CUTLASS and Tensor Core KernelsCUTLASS provides components for building GPU matrix kernels. It is useful when the required precision, batching or fused operations need more control than a library call offers.
Advanced
⚡ Kernels & Compilers🔒 Premium
GEMM Work Partitioning and Wave QuantizationOutput tiles determine how much independent matrix work is available. Split-K and work-centric schedules can reduce an underfilled final wave, while adding partial-result reduction and workspace costs.
Foundational
🧮 Open Weights & Serving Engines
Capacity Planning for Open-Weights FleetsMatch forecast traffic to a measured replica operating envelope, then check failure domains, rollout capacity, readiness time and cost. Output tokens alone omit prompt work and live state; total parameter count alone does not determine a replica's throughput.
Foundational
🧮 Open Weights & Serving Engines
Reading config.json to Size a Model You Have Never RunRead a model configuration as an architecture description, then combine it with tensor metadata and engine allocation rules. Learn which fields determine attention state, why mixed precision needs tensor-level accounting, and how to turn a first estimate into a testable per-GPU capacity plan.
Explore the full curriculum →
UP NEXT ON YOUR JOURNEY
Next in this trackA take-home gives you a working layernorm kernel at a tenth of memory bandwidth. Make it fast and justify every change.Next in this trackImplement the verification step of speculative decoding, including the rollback of the KV cache after a rejection.Next in this trackHow do you overlap host-to-device transfers with compute, and what has to be true before the overlap actually happens?
FEDITOR'S NOTE

Explain the supported-library versus customizable-kernel trade without claiming that grouped operations or epilogues are exclusive to CUTLASS. Scope byte and scheduling estimates before using them to justify custom code.

DISCUSSION · 0

No comments yet — be the first to share your approach.