Loading...
Loading...
Found 2,256 Skills
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.
Use whenever the user mentions LLM prompt/prefix cache misses, cached_tokens=0, cache_read_input_tokens/cache_creation_input_tokens, prompt_cache_key, cache_control/cachePoint placement, stable prefixes, tool/schema stability, TTFT/prefill latency, OpenAI/Claude/Bedrock/OpenRouter routing, vLLM/SGLang KV reuse, or LLM cost/speed regressions on repeated long prompts. Use when reviewing LLM request shape changes: prompt text, message order, request builders, tools, schemas, response_format, provider API surface, model/router settings, agent loop structure, context compaction, or inference deployment. Use for speeding up agents only when prompt-cache stability, TTFT, or cache cost is central. Do not use for generic prompt writing, generic RAG design, token counting, or non-LLM performance.
SQL and Python-based employee performance analytics with KPI aggregation, departmental insights, and HR dashboard generation
Guardião da arquitetura de software no SynkOS. Use esta skill quando o usuário pedir para propor ou revisar a arquitetura de um sistema, avaliar tradeoffs entre tecnologias ou abordagens, criar um ADR (Architecture Decision Record), desenhar um modelo de dados ou contrato de API, ou fazer perguntas como "qual stack usar para X?", "como estruturar esse serviço?", "quais são os tradeoffs de Y vs Z?", "documente as decisões técnicas", "revise essa arquitetura". Ative também para discovery brownfield (entender o que já existe antes de propor mudanças), para cross-cutting concerns como segurança e performance, e para revisar designs propostos pelas equipes de implementação.
Writes, reviews, and debugs idiomatic Rust code with memory safety and zero-cost abstractions. Implements ownership patterns, manages lifetimes, designs trait hierarchies, builds async applications with tokio, and structures error handling with Result/Option. Use when building Rust applications, solving ownership or borrowing issues, designing trait-based APIs, implementing async/await concurrency, creating FFI bindings, or optimizing for performance and memory safety. Invoke for Rust, Cargo, ownership, borrowing, lifetimes, async Rust, tokio, zero-cost abstractions, memory safety, systems programming.
Interpret Apache Doris query runtime profiles, especially profile bottleneck triage, misleading wait counters, per-operator metric priority, scan, join-order/runtime-filter analysis, and evidence-bounded performance explanations. Use when given a Doris profile, query id, profile URL/text, or a request to explain Doris query performance.
Rank outreach campaigns by real revenue impact — which campaigns actually generated deals, pipeline, or meetings — by cross-referencing the user's La Growth Machine campaign data with their CRM deal data (HubSpot today). Use whenever the user wants to know which campaigns drove pipeline, compare campaign ROI, see which campaigns to continue / stop / adapt, audit campaign impact, review attribution, asks 'which of my campaigns is actually working', or wants a campaign performance ranking by deals or revenue. Triggers on: 'which campaigns drove pipeline', 'rank my campaigns by deals', 'campaign ROI', 'campaign impact', 'which campaigns to stop', 'which to scale', 'attribution review', 'pipeline by campaign'. Pulls live data from the La Growth Machine MCP and the HubSpot MCP when connected; works from pasted exports otherwise. For RevOps, Heads of Sales/Marketing, founders and growth leads doing campaign performance reviews. Maintained by La Growth Machine.
Expert Django backend development guidance. Use when creating Django models, views, serializers, or APIs; debugging ORM queries or migrations; optimizing database performance; implementing authentication; writing tests; or working with Django REST Framework. Follows Django best practices and modern patterns.
Diagnose and audit SEO issues affecting crawlability, indexation, rankings, and organic performance. Use when the user asks for an SEO audit, technical SEO review, ranking diagnosis, on-page SEO review, meta tag audit, or SEO health check. This skill identifies issues and prioritizes actions but does not execute changes. For large-scale page creation, use programmatic-seo. For structured data, use schema-markup.
Compress large language models using knowledge distillation from teacher to student models. Use when deploying smaller models with retained performance, transferring GPT-4 capabilities to open-source models, or reducing inference costs. Covers temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.
Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile performance with TensorBoard - Google's ML visualization toolkit
Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance.