Data Science & ML Claude 技能

Quantitative Backtesting Frameworks

Builds robust, production-grade backtesting systems to validate trading strategies while eliminating common statistical biases.

MLOps Pipeline & Workflow Orchestration

Builds and automates end-to-end MLOps pipelines from data preparation and model training to production deployment and monitoring.

LLM Evaluation Framework

Implement comprehensive evaluation strategies for LLM applications using automated metrics, human feedback, and benchmarking.

RAG Implementation Suite

Builds robust Retrieval-Augmented Generation systems using vector databases, semantic search, and advanced retrieval patterns for LLM applications.

Vector Index Tuning Specialist

Optimizes vector database performance by tuning HNSW parameters, quantization strategies, and memory usage for efficient AI applications.

Advanced Prompt Engineering Patterns

Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production environments.

Embedding Strategies for RAG

Optimizes embedding model selection and chunking strategies to improve semantic search and RAG application performance.

ESM Protein Design & Modeling

Designs, analyzes, and generates protein sequences and structures using Evolutionary Scale Modeling (ESM3 and ESM C).

Trust RL Model Predictions

Optimizes trading strategy execution by removing redundant heuristic pattern filters that conflict with Reinforcement Learning model signals.

Spark Performance Optimizer

Optimizes Apache Spark jobs through advanced partitioning, memory management, and shuffle tuning to improve performance and scalability.

ADK Agent Engineer

Builds and deploys production-ready Agent Development Kit (ADK) agents with robust testing and multi-agent orchestration.

Google Veo Video Generation

Generates high-fidelity videos and synchronized audio using Google Veo 3.1 via the Vertex AI API.

Multiplex Imaging Analysis

Optimizes segmentation, feature extraction, and spatial analysis workflows for high-dimensional multiplex immunofluorescence imaging data.

Python Dataclass Optimizer

Optimizes Python dataclasses for memory efficiency, immutability, and validation using advanced PEP 557 patterns.

Excel Data Extraction & Transformation

Extracts, structures, and converts complex Excel data into developer-friendly JSON and CSV formats.

Data Analyst Export

Exports analysis results and query data into professional formats including CSV, JSON, Excel, and Markdown tables.

RL Trading Reward Function Optimizer

Eliminates HOLD bias in reinforcement learning trading models by calibrating reward functions and slippage penalties.

Gemini AI Integration

Delegates complex tasks and analysis to Google's Gemini models directly within your Claude Code environment.

RLM Reasoning for Large Context

Optimizes Claude's ability to process massive files and complex codebases using a recursive Probe-Extract-Confirm reasoning pattern.

Quantitative Design Consultant

Architects optimized quantitative research designs, experimental methodologies, and sampling strategies using an enhanced three-phase validation process.

Scanner PDF Analysis

Extracts tables, identifies sections, and summarizes content from complex PDF documents and technical reports.

Qualitative Research Coding Specialist (E2)

Automates and guides systematic qualitative data coding, theme development, and rigorous research methodology implementation.

Trading Prediction Outcome Tracker

Automates the recording of actual trading results against predicted signals to enable accurate performance metrics and model retraining.

I3 RAG Builder & Document Processor

Builds cost-free RAG systems using parallel document processing and local vector embeddings.

Multi-Timeframe RL Training

Trains Reinforcement Learning models across multiple market timeframes with automated data resampling and professional market alignment.

PRISMA Systematic Review Orchestrator

Automates and manages the end-to-end 7-stage PRISMA 2020 systematic literature review pipeline from research question to RAG system.

Quarto Anti-Slop

Enforces reproducible research and academic writing standards for Quarto and RMarkdown documents to eliminate generic AI-generated content.

AgentDB Reinforcement Learning

Implements and trains autonomous agents using nine specialized reinforcement learning algorithms for self-improving behavior.

Flow Nexus Neural

Train, deploy, and manage distributed neural networks across sandboxed E2B environments using multiple architectures and federated learning.

Macro Fiscal & Labor Scenario Analyzer

Analyzes US fiscal deficit expansion and Treasury risk scenarios during periods of rising unemployment and high GDP growth.