cursor-croq-tune

Orchestrate GPU kernel autotuning cycles for CUDA and DSL-targeted kernels on NVIDIA GPUs.

Updated Apr 10, 2026

Search Tags:#deep learning #performance profiling #cuda #kernel optimization #automated tuning #gpu tuning

Ready to run

npx skills add https://github.com/LancerLab/croqtile-tuner --skill cursor-croq-tune

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: cursor-croq-tune
Source: https://github.com/LancerLab/croqtile-tuner/tree/main/.cursor/skills/cursor-croq-tune
Command: npx skills add https://github.com/LancerLab/croqtile-tuner --skill cursor-croq-tune

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the process of tuning GPU kernels, including build, profiling, and optimization, to achieve peak performance without manual intervention.

Core Features & Use Cases

End-to-End Tuning Workflow: Coordinates building, profiling, ideation, implementation, verification, measurement, and decision-making sequences.
Automated Data Handling: Manages results storage, checkpointing, state validation, and resume capabilities for long tuning sessions.
Use Case: For a developer optimizing a new matrix multiplication kernel, this Skill can run the entire tuning loop, rapidly finding and validating high-performance configurations.

Quick Start

Load this Skill and set parameters for your GPU, specific kernel shape, and object. Then initiate tuning with your desired dataset and target architecture.

Frequently Asked Questions about cursor-croq-tune

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I automate GPU kernel tuning for CUDA?▼

Automate GPU kernel tuning by orchestrating build, profile, and optimization cycles for CUDA kernels. This coordinates the entire tuning loop, rapidly finding and validating high-performance configurations without manual intervention.

What is the best way to profile and optimize deep learning kernels on NVIDIA GPUs?▼

The best way to optimize deep learning kernels is running an end-to-end tuning workflow that handles profiling, ideation, implementation, verification, and measurement to find peak performance configurations on NVIDIA GPUs.

Do I need specific environment setup tools for automated GPU tuning?▼

Yes, automated GPU tuning requires an environment setup including GPU, compiler, and profiling tools. These prerequisites are necessary to execute the build, profile, and optimization cycles for your kernels.

Can I resume long kernel optimization sessions after a crash?▼

Yes, you can resume long kernel optimization sessions after a crash using automated data handling. The process manages results storage, checkpointing, state validation, and resume capabilities for continuous tuning.

How does automated kernel tuning handle large model training tasks?▼

Automated kernel tuning handles large model training tasks by applying deep performance tuning to DSL-targeted kernels. It coordinates efficient search and validation to ensure optimal kernel configurations for demanding training workloads.