How does it handle model interpretability?

The skill includes patterns for computing SHAP values to validate feature importance and ensure model decisions align with business logic.

Is causal inference supported?

Yes, it provides implementation patterns for Difference-in-Differences (DiD) modeling with robust standard errors and parallel trend validation.

Does it provide experiment design guidance?

Absolutely. It provides sample size calculators, two-proportion z-tests, and rigorous checklists to avoid sample ratio mismatch and data leakage.

Can it help with imbalanced datasets?

Yes, it includes workflows for AUC-PR evaluation and stratified cross-validation specifically designed for imbalanced classification tasks.

What libraries does this skill support?

The skill is optimized for Python (NumPy, Pandas, Scikit-learn, XGBoost, SciPy, Statsmodels), R, and SQL.

Senior Data Scientist

Name: Senior Data Scientist
Author: zhangzhang-111-i

byzhangzhang-111-i

0•

Ciencia de Datos y ML

Implements production-grade statistical modeling, experiment design, and predictive analytics pipelines using Python, R, and SQL.

This skill transforms Claude into a senior data scientist capable of architecting end-to-end machine learning and statistical workflows. It provides specialized guidance for designing rigorous A/B tests with power analysis, building robust feature engineering pipelines with Scikit-learn, and performing advanced causal inference using Difference-in-Differences. It is particularly useful for engineering teams needing to move beyond basic data scripts into production-ready MLflow tracking, cross-validated model evaluation, and business-driven data analysis that accounts for statistical significance and data leakage.

Características Principales

01Production-grade feature engineering pipelines with leakage prevention and time-series logic

02Automated A/B test design including sample size calculations and significance testing

030 GitHub stars

04Comprehensive model evaluation using AUC-PR, SHAP, and stratified cross-validation

05MLflow integration for experiment tracking and lifecycle management

06Advanced causal inference modeling via Difference-in-Differences (DiD) and OLS

Casos de Uso

01Performing causal analysis on observational data to estimate treatment effects and parallel trends

02Building scalable machine learning pipelines for classification or regression tasks on structured data

03Designing and analyzing controlled experiments to measure the impact of product changes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zhangzhang-111-i/claude-skills111 senior-data-scientist

For use in Claude.ai and ChatGPT

Características Principales

01Production-grade feature engineering pipelines with leakage prevention and time-series logic

02Automated A/B test design including sample size calculations and significance testing

030 GitHub stars

04Comprehensive model evaluation using AUC-PR, SHAP, and stratified cross-validation

05MLflow integration for experiment tracking and lifecycle management

06Advanced causal inference modeling via Difference-in-Differences (DiD) and OLS

Casos de Uso

01Performing causal analysis on observational data to estimate treatment effects and parallel trends

02Building scalable machine learning pipelines for classification or regression tasks on structured data

03Designing and analyzing controlled experiments to measure the impact of product changes