optimization▌

Iteratively optimize cuTile kernel performance through systematic profiling, bottleneck analysis, IR comparison, and targeted tuning. Covers tile sizes, occupancy, autotune configs, TMA, latency hints, persistent scheduling, num_ctas, flush_to_zero, and IR-level debugging. Use when asked to "optimize cutile kernel", "improve kernel perf", "tune cutile performance", "make kernel faster", or iteratively benchmark and refine a cuTile GPU kernel in the TileGym project.

Optimizes application performance by identifying and fixing bottlenecks based on profiling data.

Guides GEO/AEO strategy for AI search visibility. GEO optimizes content for ChatGPT, Claude, Perplexity, and AI search summaries (Google AI Overviews, Bing Copilot, Yandex Search with AI)—getting cited in AI-generated answers rather than ranking in traditional SERPs. See serp-features for AI search as SERP features; featured-snippet for snippet optimization that overlaps with AI Overviews.

Complete App Store Optimization toolkit for researching, optimizing, and tracking mobile app performance across Apple App Store and Google Play Store. \n \n Covers keyword research, competitor analysis, metadata optimization (titles, descriptions, keywords), and visual asset strategies with platform-specific character limit validation \n Includes review sentiment analysis, rating improvement tactics, and A/B testing frameworks for continuous conversion optimization \n Provides ASO health scoring

Identify and quantify Azure cost savings through resource analysis, utilization metrics, and actionable optimization recommendations. \n \n Discovers orphaned resources (unattached disks, unused NICs, idle gateways) and over-provisioned services using Azure Quick Review scans \n Queries actual costs from Azure Cost Management API and utilization data from Azure Monitor to support rightsizing recommendations \n Generates prioritized optimization reports with estimated savings, validated pricing,

Production patterns for optimizing Apache Spark jobs including partitioning strategies, memory management, shuffle optimization, and performance tuning.

Production patterns for Bazel in large-scale monorepos.

Strategies and patterns for optimizing cloud costs across AWS, Azure, and GCP.

Reduce infrastructure costs through intelligent resource allocation, reserved instances, spot instances, and continuous optimization without sacrificing performance.

LLM applications emit spans following OpenInference semantic conventions. Prompts are stored in different span attributes depending on the span kind and instrumentation: