Data Science & ML Claude 技能

DSPy Bootstrap Few-Shot Optimizer

Automates the generation and selection of high-quality few-shot demonstrations for DSPy programs using teacher models.

DSPy MIPROv2 Optimizer

Optimizes DSPy programs by jointly tuning instructions and few-shot demonstrations using Bayesian search for maximum model performance.

DSPy ReAct Agent Builder

Builds production-ready ReAct agents using the DSPy framework to solve complex multi-step tasks with integrated tool-calling and reasoning.

DSPy SIMBA Optimizer

Optimizes DSPy programs using mini-batch Bayesian optimization and statistical analysis of rich feedback signals.

DSPy Signature Designer

Designs type-safe, structured signatures for DSPy modules to define precise AI model inputs and outputs.

DSPy Output Refinement & Constraints

Enhances DSPy program reliability by enforcing custom constraints and reward functions through iterative refinement and best-of-N selection.

DSPy Advanced Module Composition

Orchestrates complex DSPy programs using ensemble patterns, multi-chain reasoning synthesis, and robust sequential pipelines.

DSPy GEPA Reflective Optimizer

Optimizes complex AI agents and ReAct systems using LLM-driven reflection on execution trajectories and textual feedback.

DSPy RAG Pipeline Builder

Builds and optimizes retrieval-augmented generation pipelines using the DSPy framework and ColBERTv2 for grounded, factual AI responses.

DSPy Custom Module Designer

Architects production-grade custom DSPy modules with robust state management, serialization, and error handling.

DSPy Evaluation Suite

Systematically measures and evaluates DSPy program performance using built-in metrics and custom scoring functions.

Excel Spreadsheet Management & Analysis

Automates the creation, editing, and analysis of professional Excel spreadsheets with advanced formula support and financial modeling standards.

Venue Expert

Provides deep expertise on global exchange mechanics and market microstructure for building and debugging trading systems.

Prompt Engineering

Creates and optimizes advanced prompts using patterns like few-shot learning, chain-of-thought, and system prompt design to significantly improve LLM performance.

ML-Ralph Project Architect

Generates comprehensive Product Requirements Documents (PRDs) for machine learning experiments and Kaggle competitions through guided conversation.

Semantic Embeddings

Generates dense vector representations for text to enable semantic search, document similarity, and RAG pipelines.

RAG Retrieval Patterns

Implements production-ready Retrieval-Augmented Generation (RAG) patterns to ground LLM responses in verifiable context and prevent hallucinations.

Reranking Patterns for RAG

Optimizes search precision in RAG pipelines by re-scoring retrieved documents using cross-encoders, LLMs, and weighted relevance signals.

Signal Correlation Workbench

Correlates qualitative customer feedback with quantitative telemetry and revenue data to drive data-backed business decisions.

LangGraph Supervisor Pattern

Coordinates complex multi-agent workflows using a central supervisor to route tasks among specialized LangGraph workers.

Golden Dataset Validator

Ensures the integrity and quality of AI evaluation datasets through automated schema validation, duplicate detection, and coverage analysis.

LangGraph Routing Patterns

Implements dynamic, state-based routing and retry logic for complex LangGraph AI agent workflows.

LangGraph Human-in-the-Loop

Implements robust human-in-the-loop patterns for LangGraph workflows to enable manual review gates and approval-based agent supervision.

Ollama Local Inference

Enables high-performance local LLM execution for cost-effective, private, and offline AI-powered development.

LangGraph State Management

Implements robust state management patterns for LangGraph workflows using TypedDict, Pydantic, and custom reducers.

Multi-Agent Orchestration

Coordinates multiple specialized AI agents using architectural patterns like fan-out/fan-in, supervisors, and synthesis for complex task execution.

Query Decomposition

Decomposes complex multi-concept queries into independent searchable terms to optimize retrieval accuracy and coverage in RAG pipelines.

LangGraph Functional API

Builds complex AI workflows using decorator-based patterns for parallel execution, persistence, and human-in-the-loop interactions.

Semantic Caching for LLMs

Optimizes LLM performance and reduces API costs by implementing Redis-powered semantic similarity caching.

Golden Dataset Management

Protects and maintains high-quality test datasets for AI/ML systems through automated backup, restoration, and integrity validation.