Best Claude prompts for Operations Research Analysts

Given a dataset of 10 million customer records with 50 variables, and an existing ETL pipeline written in Python using the pandas library, identify the bottlenecks in the current pipeline and propose a revised pipeline that utilizes Snowflake for data warehousing and Tableau for data visualization. Assume the pipeline runs daily and the data is sourced from a SQL database. Provide a detailed step-by-step plan for implementation, including code snippets and expected performance improvements.

Using a sample dataset of 100,000 observations, compare the precision of three different machine learning models (linear regression, decision tree, and random forest) in predicting customer churn. Implement each model in R and calculate the mean absolute error, mean squared error, and R-squared value for each model. Provide a written summary of the results, including visualizations of the residuals and a discussion of the implications for business stakeholders.

Given a complex SQL query that joins five tables and performs aggregations on multiple columns, analyze the query plan and identify opportunities for optimization to improve data accuracy and reduce query execution time. Propose a rewritten query that utilizes efficient indexing strategies and minimizes data redundancy, and provide a step-by-step explanation of the optimization process.

Using a pre-trained regression model that predicts employee salaries based on demographic and performance variables, interpret the model coefficients and identify the most influential predictors. Refine the model by selecting a subset of the most important features and retraining the model, then compare the performance of the original and refined models using metrics such as mean squared error and R-squared. Provide a written report of the results, including visualizations of the partial dependence plots and a discussion of the implications for HR stakeholders.