Data Science & ML Claude 스킬

D3.js Data Visualization

Creates sophisticated, interactive data visualizations and custom charts using the D3.js library.

Forecasting Calibration & Discrimination

Optimizes predictive accuracy by balancing probability alignment with the ability to distinguish between diverse outcomes.

Natural Selection Design Patterns

Implements evolutionary design principles to build adaptive systems that improve through iterative variation, selection pressure, and inheritance.

Deep Research Oracle

Routes prompts and code context to multiple high-performance AI models for cross-validation and specialized deep research.

Prompt Engineering Frameworks

Applies systematic techniques and structured frameworks to optimize LLM instructions for maximum accuracy, consistency, and output quality.

Evolution Patterns

Applies biological adaptation principles to optimize complex systems through iterative variation, selection, and inheritance.

Base Rate Thinking Pattern

Anchors predictions and decision-making in statistical frequencies to avoid cognitive bias and improve estimation accuracy.

Aristotle Mathematical Prover

Formalizes natural language mathematical questions into Lean 4 and verifies them using the Harmonic Aristotle prover API.

Tournament Forecasting

Implements structured competitive prediction frameworks using Brier scores and systematic debiasing to enhance organizational forecasting accuracy.

ctrlsys Control Theory & Systems Engineering

Simplifies the design and analysis of complex control systems using the C11-based ctrlsys library.

Fitness Landscapes Thinking Pattern

Analyzes complex optimization problems using evolutionary landscape metaphors to identify local traps and global optima.

Excel & Spreadsheet Management

Manages complex Excel workbooks with automated formula creation, financial modeling standards, and data analysis.

Stable Baselines3 Reinforcement Learning

Implements and optimizes reinforcement learning workflows using the Stable Baselines3 PyTorch library.

Trading Risk Sizing Optimizer

Refactors trading system logic to transform rigid trade rejections into intelligent, constraint-based position sizing.

Confidence Intervals Thinking Pattern

Quantifies uncertainty in estimates by generating plausible ranges to enable more reliable data-driven decision making.

Autocatalytic Sets Mental Model

Analyzes and designs self-sustaining systems through the lens of collective catalysis and network closure.

Bayes' Theorem Analysis

Refines probability estimates and decision-making by systematically updating beliefs as new data or evidence emerges.

Fireworks AI Inference & Fine-tuning

Integrates high-performance inference and LoRA fine-tuning for 100+ open-source LLMs via OpenAI-compatible APIs and the firectl CLI.

Batch Serving ML Pattern

Implements high-throughput machine learning inference patterns for processing large-scale datasets on a scheduled basis.

Fat Tails Risk Analysis

Identifies and manages risks in fat-tailed distributions where extreme events occur more frequently than standard models predict.

Extremizing Forecasts

Adjusts and optimizes aggregated probability forecasts to compensate for crowd conservatism and improve predictive accuracy.

AI Agents Architect

Architects autonomous AI agent systems using structured reasoning patterns, memory management, and resilient tool integration.

Brier Score Forecasting Accuracy

Quantifies the accuracy of probabilistic predictions to improve decision-making and calibration through systematic feedback.

Pareidolia Pattern Analysis

Prevents false positive pattern recognition in data and visual analysis by distinguishing genuine signals from cognitive illusions.

ML Batch Processing Pattern

Implements a decoupled architecture for pre-computing machine learning predictions at scheduled intervals to optimize costs and serving latency.

Megatrends Analysis

Identifies long-term societal and structural shifts through bottom-up pattern detection and massive data aggregation.

RAG Engineer

Optimizes Retrieval-Augmented Generation architectures through advanced semantic chunking, hybrid search strategies, and vector embedding pipelines.

Computational Mechanics

Identifies the minimal computational structure and causal states needed to predict complex system behavior from observed data streams.

Monte Carlo Simulation

Quantifies uncertainty and assesses risk distributions by running thousands of probabilistic scenarios with random variable inputs.

Bayesian Updating

Refines confidence in hypotheses and technical decisions by systematically weighting prior beliefs against new evidence.