Data Science & ML Agent Skills

LLM Evaluation Framework

Implements rigorous evaluation strategies for LLM applications using automated metrics, human-in-the-loop feedback, and systematic benchmarking.

Claude Batch Processing

Optimizes high-volume workloads by leveraging Anthropic's Message Batches API for 50% cost savings on non-time-sensitive tasks.

Macro Trend Deviation Analyzer

Analyzes asset price deviations from long-term exponential trends to identify historical extremes and macro market regimes.

Kalshi Prediction Markets

Accesses real-time prediction market data, betting odds, and trading analytics from the Kalshi platform.

DSPy Declarative Prompting

Optimizes AI system prompts automatically using the DSPy framework to build modular, data-driven LLM pipelines.

Trading Model Post-Training Workflow

Validates and assesses reinforcement learning trading models through systematic gating, backtesting, and walk-forward validation.

Advanced Excel & Spreadsheet Manager

Manages complex Excel workbooks with automated formula recalculation, professional financial modeling standards, and deep data analysis capabilities.

Dask Parallel Computing

Scales Python data workflows across multiple cores or clusters to handle datasets exceeding available memory.

Prompt Engineering Patterns

Implements advanced prompt engineering techniques to optimize LLM performance, reliability, and structured output in production environments.

ML Model Evaluation Suite

Evaluates machine learning models by generating comprehensive performance metrics including accuracy, precision, and F1-score to guide model optimization.

Prompt Engineering & RAG Guide

Optimizes LLM performance through advanced prompt engineering, RAG system design, and agent workflow orchestration.

Agent Workflow Patterns

Implements sophisticated autonomous agent architectures and workflow patterns using the Vercel AI SDK.

Multi-Agent Architecture Patterns

Architects complex LLM workflows using advanced multi-agent patterns like supervisors, swarms, and hierarchical delegation for optimized context management.

LangChain Application Architect

Designs and implements sophisticated LLM applications using LangChain's framework for agents, memory, and complex workflows.

Data Anomaly Detection

Identifies outliers and unusual patterns in datasets using advanced machine learning algorithms to uncover fraud, defects, or security threats.

RAG Implementation Patterns

Implements high-performance Retrieval Augmented Generation (RAG) pipelines including document chunking, vector database integration, and semantic search.

LLM Application Evaluation

Implements comprehensive evaluation frameworks for LLM applications using automated metrics, human feedback, and systematic benchmarking.

TimeGPT-2 Migration Assistant

Automates the transition of time-series forecasting pipelines from TimeGPT-1 to the enhanced TimeGPT-2 architecture.

RAG Retrieval Patterns

Implements and optimizes advanced search strategies including semantic, hybrid, and reranking for high-performance RAG systems.

AI & ML Prompt Optimizer

Optimizes Large Language Model (LLM) prompts to minimize token consumption, reduce operational costs, and enhance response quality.

Document Parsing & Extraction

Extracts text, tables, and metadata from PDF, DOCX, and HTML documents to power RAG pipelines and data processing workflows.

ElevenLabs Speech-to-Text Integration

Integrates ElevenLabs Scribe v1 for high-accuracy speech-to-text transcription across 99 languages with speaker diarization.

Multimodal Models (Vision, Audio & Image Gen)

Integrates pre-trained models like CLIP, Whisper, and Stable Diffusion for advanced vision, speech recognition, and image generation tasks.

Gemma 3 270M Integration Template

Streamlines the integration and testing of Gemma 3 270M models within the Claude Code environment.

Code Analyzer

Analyzes GitHub repositories to extract computational methodologies and automatically draft scientific Methods sections.

Spark Fundamentals & PySpark

Streamlines distributed data processing by providing standardized PySpark patterns and performance best practices.

Prompt Engineering Patterns

Optimizes LLM performance and reliability through advanced prompting techniques like few-shot learning and chain-of-thought reasoning.

Scientific Visualization & Figure Design

Generates publication-ready scientific figures and multi-panel layouts optimized for top-tier academic journals like Nature and Science.

LlamaIndex RAG Patterns

Implements production-ready RAG pipelines and advanced retrieval strategies using LlamaIndex templates and scripts.

Mem0 Memory Design Patterns

Architects scalable AI memory systems with optimized retention, storage backends, and multi-level context patterns.