Total 32,819 skills
Showing 12 of 32819 skills
Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision/offload config, and distributed checkpointing. Use when models exceed single-GPU memory or when you need DTensor-based sharding with DeviceMesh.
React component and hook testing patterns with Testing Library and Vitest. Use when writing tests for React components, custom hooks, or data fetching logic. Covers component rendering tests, user interaction simulation, async state testing, MSW for API mocking, hook testing with renderHook, accessibility assertions, and snapshot testing guidelines. Does NOT cover E2E tests (use e2e-testing) or TDD workflow enforcement (use tdd-workflow).
Research high-performing Instagram content (posts and reels) from tracked accounts using Apify's Instagram Scraper. Identifies outlier content, analyzes top 5 videos with AI, and generates reports with actionable hook formulas. Use when asked to: - Find trending Instagram content in a niche - Research what's performing on Instagram - Identify high-performing reel patterns - Analyze competitors' Instagram content - Generate content ideas from Instagram trends - Run Instagram research - Find viral reels - Analyze hooks and content structure Triggers: "instagram research", "ig research", "find trending reels", "analyze instagram accounts", "what's working on instagram", "content research instagram", "reel analysis", "instagram trends"
Scalable data processing for ML workloads. Streaming execution across CPU/GPU, supports Parquet/CSV/JSON/images. Integrates with Ray Train, PyTorch, TensorFlow. Scales from single machine to 100s of nodes. Use for batch inference, data preprocessing, multi-modal data loading, or distributed ETL pipelines.
Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan.
Provides guidance for enterprise-grade RL training using miles, a production-ready fork of slime. Use when training large MoE models with FP8/INT4, needing train-inference alignment, or requiring speculative RL for maximum throughput.
Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.
Aggregate news from popular cryptocurrency RSS feeds, analyze sentiment of articles, and calculate an overall market sentiment score with detailed explanation. Use when assessing crypto market sentiment for trading decisions, research, or monitoring trends from RSS sources.
Clean code principles, SOLID, and code review practices
Write structured product requirements documents (PRDs) with problem statements, user stories, requirements, and success metrics. Use when speccing a new feature, writing a PRD, defining acceptance criteria, prioritizing requirements, or documenting product decisions.
This skill should be used when the user asks to "create a Feishu doc", "create a new doc in Feishu", "open Feishu and create document", "create document in leiniao-ibg", or mentions creating documents in Feishu/Lark workspace.
Generate human-like speech audio with Model Studio DashScope Qwen TTS (qwen3-tts-flash). Use when converting text to speech, producing voice lines for short drama/news videos, or documenting TTS request/response fields for DashScope.