Individual skill

tilegym-cutile-autotuning

Use when adding, modifying, optimizing, or debugging CuTile autotuning code. Trigger signals: `exhaustive_search` / `replace_hints` / `hints_fn` / `cuda.tile.tune` in code, `autotune` in filenames, or correctness/performance issues in autotuned CuTile kernels. Covers: tune-once/cache/launch pattern, per-architecture configs (sm80–sm120), parameter space design (tile sizes, occupancy, num_ctas), and 7 common pitfalls with solutions.

skills/tilegym-cutile-autotuning/SKILL.md

Read pinned source

Source commit fce3cca4170be75e82c64920654d56ee2811b005

What was checked

Matches the observed repository HEAD: Yes
Scanner rules are current: Yes
Repository scan complete: No
Skill scan complete: No
Observed behavior: Network or broad shell
Last checked: Nov 9, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings): 3,906
Repository last updated: Nov 6, 2026
License: Apache-2.0

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

agent-skills Repository topic
agentic-ai Repository topic
ai-agents Repository topic
claude-code Repository topic
codex Repository topic
coding-agents Repository topic
cuda Repository topic
developer-tools Repository topic
llm Repository topic
omniverse Repository topic
engineering Derived category
design Derived category
testing Derived category

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files

BENCHMARK.mddoc · 5,399 bytes
SKILL.mdcore · 18,703 bytes
SKILL.mdcore · 18,703 bytes
SKILL.mdcore · 18,703 bytes
SKILL.mdcore · 18,703 bytes
assets/examples/01_rmsnorm_occupancy_only/autotuned_launch.pyruns · 8,084 bytes
assets/examples/01_rmsnorm_occupancy_only/fixed_launch.pyruns · 5,736 bytes
assets/examples/02_matmul_full_search/autotuned_launch.pyruns · 9,659 bytes
assets/examples/02_matmul_full_search/fixed_launch.pyruns · 5,559 bytes
assets/examples/03_rope_inplace_splitbuffer/autotuned_launch.pyruns · 11,536 bytes
assets/examples/03_rope_inplace_splitbuffer/fixed_launch.pyruns · 7,724 bytes
evals/evals.jsonasset · 5,405 bytes
references/api-reference.mddoc · 8,005 bytes
references/hardware-constraints.mddoc · 11,402 bytes
references/kernel-type-templates.mddoc · 51,640 bytes
references/parameter-space-design.mddoc · 11,164 bytes
references/pitfalls.mddoc · 5,584 bytes
references/search-strategies.mddoc · 9,024 bytes
references/workflow.mddoc · 17,136 bytes
skill-card.mddoc · 4,020 bytes
skill.oms.sigasset · 7,993 bytes

Read excerpt


# CuTile Autotuning

Add autotuning to CuTile kernels using the `exhaustive_search` API with tune-once/cache/direct-launch pattern.

## Instructions

Follow the decision tree to classify the kernel, design a search space, implement the tune-once/cache/launch pattern, and validate performance.

1. **Classify** — use the Decision Tree to determine search dimensions (occupancy-only vs full tile search)
2. **Design search space** — select the matching template from `references/kernel-type-templates.md`; prune to ≤ 30 configs in the final code via arch filters (directed exploration probes may temporarily exceed this — see Design Philosophy)
3. **Implement** — add `exhaustive_search` + cache + `ct.launch` following the Step-by-Step Workflow; handle in-place writes with split-buffer if needed
4. **Test** — run correctness with autotune enabled and with `DISABLE_AUTOTUNE=1`
5. **Validate** — A/B benchmark against fixed best-known config; see `references/search-strategies.md`
6. **Shrink** — prune dead-weight configs that never win, targeting ≤ 8 configs per architecture to minimize compilation cost (Step 10)