Ab Testing Statistics Ai Ml Stub
Design and analyze A/B tests with statistical rigor: sample sizing, significance testing, CUPED variance reduction, multi-armed bandits, and experiment reporting.
Ai Cost Optimizer
>
Skill Name
One-line description of what this skill does and when to use it.
Mirascope
Clean, type-safe LLM API wrapper with structured outputs, streaming, and provider-agnostic interface
Haystack
Production-ready LLM pipeline framework with component-based architecture for RAG and QA
Huggingface Datasets
Use this skill for Hugging Face Dataset Viewer API workflows that fetch subset/split metadata, paginate rows, search text, apply filters, download parquet URLs, and read size or statistics.
Guidance
Microsoft's structured generation library for constrained LLM outputs using Handlebars-style templates
Modal Gpu
Serverless GPU compute for ML workloads — run training, inference, and batch jobs on Modal
Huggingface Local Models
Use to select models to run locally with llama.cpp and GGUF on CPU, Mac Metal, CUDA, or ROCm. Covers finding GGUFs, quant selection, running servers, exact GGUF file lookup, conversion, and OpenAI-compatible local serving.
Ollama Integration
>
Dspy
Build complex AI systems with declarative programming, optimize prompts automatically, create modular RAG systems and agents with DSPy - Stanford NLP's framework for systematic LM programming
Peft Fine Tuning
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.
Marvin
AI functions and classifiers using LLMs as pure Python functions with type annotations
Kor
Structured data extraction from LLMs using schema definitions and few-shot examples
Ollama Python Sdk
Programmatic local LLM usage via the official Ollama Python SDK
Chroma
Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source projects.
Prompt Chaining
Build reliable multi-step LLM pipelines using prompt chaining — decomposing complex tasks into sequential prompts where each output feeds the next. Covers gate conditions, parallel chains, and error handling.
Mlflow Experiment Tracking
Track ML experiments, manage model versions, and deploy models using MLflow. Covers experiment logging, model registry, artifact management, and serving patterns.
Langgraph
Build stateful multi-actor AI applications with LangGraph: cyclic graphs, agent workflows, fine-grained state/flow/memory control.
Autogen
Microsoft's multi-agent conversation framework for building collaborative AI workflows
Mteb
Massive Text Embedding Benchmark for evaluating and comparing embedding models across tasks
Outlines
Fast structured text generation from LLMs using regex and JSON schema constraints
Llm Fine Tuning Pipeline
Build end-to-end LLM fine-tuning pipelines — dataset preparation with quality filtering, SFT training with Hugging Face Trainer and TRL, LoRA/QLoRA PEFT configuration, DPO preference optimization, evaluation with LM-Eval, and vLLM serving of fine-tuned adapters.
Langchain
Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), 500+ integrations, ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG applications. Best for rapid prototyping and production deployments.
Embeddings Expert
>
Pinecone
Managed vector database for production AI applications. Fully managed, auto-scaling, with hybrid search (dense + sparse), metadata filtering, and namespaces. Low latency (<100ms p95). Use for production RAG, recommendation systems, or semantic search at scale. Best for serverless, managed infrastructure.
Openrouter Litellm
>
Faiss
Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU acceleration, and various index types (Flat, IVF, HNSW). Use for fast k-NN search, large-scale vector retrieval, or when you need pure similarity search without metadata. Best for high-performance applications.
Multi Agent Orchestration
Design and implement multi-agent systems where specialized AI agents collaborate to complete complex tasks. Covers orchestrator-worker patterns, parallel execution, result aggregation, error recovery, agent handoffs, and production deployment with LangGraph and CrewAI.
Llm Evaluation
Master comprehensive evaluation strategies for LLM applications, from automated metrics to human evaluation and A/B testing.
Embedding Pipeline
Build production embedding pipelines for semantic search, RAG, and recommendation systems. Covers text chunking strategies, batch embedding with OpenAI/Anthropic/HuggingFace, vector storage, embedding updates, and monitoring embedding quality.
Knowledge Graph
Build and query knowledge graphs using Neo4j, NetworkX, and RDF/SPARQL for representing complex relationships. Covers graph modeling, Cypher queries, graph embeddings, community detection, and integrating knowledge graphs with LLM RAG pipelines.
Multimodal Ai
>
Computer Vision
>
Claude Tool Use
>
Litellm Proxy
Deploy LiteLLM as a unified OpenAI-compatible API gateway for 100+ LLM providers
Llm Streaming
>
Instructor
Extract structured data from LLM responses with Pydantic validation, retry failed extractions automatically, parse complex JSON with type safety, and stream partial results with Instructor - battle-tested structured output library
Llm Caching Strategies
Implement prompt caching, semantic caching, and response memoization for LLM applications to reduce latency and API costs by 50-90%. Covers Anthropic prompt caching, Redis semantic cache, exact-match caching, and cache invalidation strategies.
Llamafile
Run LLMs as self-contained single-file executables with no installation required
Ml Feature Engineering
Design and implement ML feature engineering pipelines — numerical and categorical encoding, text/embedding features, temporal features, feature selection with mutual information and SHAP, scikit-learn pipelines, and feature store serving for training/inference consistency.
Llm Prompt Optimizer
Use when improving prompts for any LLM. Applies proven prompt engineering techniques to boost output quality, reduce hallucinations, and cut token usage.
Crewai Multi Agent
Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents working together on complex tasks, when you need role-based agent collaboration with memory, or for production workflows requiring sequential/hierarchical execution. Built without LangChain dependencies for lean, fast execution.
Ai Agent Memory
Design and implement memory systems for AI agents — short-term working context, episodic memory in vector stores, entity extraction into knowledge graphs, procedural memory for learned behaviors, and memory consolidation. Covers LangGraph memory, Mem0, and custom multi-tier architectures.
Letta Memgpt
Build LLM agents with persistent memory and stateful long-term context (formerly MemGPT)
Hf Cli
Hugging Face Hub CLI (`hf`) for downloading, uploading, and managing models, datasets, spaces, buckets, repos, papers, jobs, and more on the Hugging Face Hub. Use when: handling authentication; managing local cache; managing Hugging Face Buckets; running or scheduling jobs on Hugging Face infrastructure; managing Hugging Face repos; discussions and pull requests; browsing models, datasets and spa…
Diffusion Models
Build and deploy diffusion model pipelines for image generation, inpainting, and fine-tuning using Hugging Face Diffusers. Covers Stable Diffusion, SDXL, ControlNet, LORA fine-tuning, img2img, inpainting, and production serving with batching.
Multimodal Processing
Build pipelines for processing multimodal data: image+text, audio+text, video analysis, cross-modal embeddings, and fusion architectures.