Best Claude prompts for Database Administrators

Given a dataset with 10 million rows, containing customer demographic information, and an ETL pipeline built using Python and Snowflake, with 5 transformation steps, identify the 3 most critical performance bottlenecks in the pipeline, and provide a step-by-step plan to optimize the pipeline, including code snippets in Python and SQL, to reduce the overall processing time by 30% while maintaining data accuracy.

Analyze a slow-running SQL query that joins 3 tables, each with 1 million rows, and provides a statistical summary of customer purchase behavior, and identify the top 2 causes of performance degradation, providing a detailed explanation of the root cause, and recommend a set of indexing strategies and query rewriting techniques to improve the query performance by 50%, including example code snippets in SQL and Python.

Evaluate the performance of a regression model built using Python and scikit-learn, trained on a dataset with 100,000 rows, containing customer transactional data, and provide a detailed analysis of the model's precision, including a comparison of actual vs. predicted values, and identify the top 3 factors contributing to model error, recommending a set of techniques to improve model precision, including feature engineering and hyperparameter tuning, with example code snippets in Python.

Perform a comprehensive analysis of data quality issues in a dataset with 50,000 rows, containing customer contact information, and identify the top 2 data quality issues, including data inconsistency and data incompleteness, providing a detailed report on the root cause of each issue, and recommend a set of data cleaning and validation techniques to improve data accuracy, including example code snippets in Python and SQL, and a statistical summary of the improved data quality.