Prompt:

"Write an SQL query to retrieve the names and email addresses of all customers who made a purchase in the last month. Tables: customers (customer_id, name, email) and orders (order_id, customer_id, order_date, amount)."

Prompt:

"I need to join three tables: customers, orders, and products. Customers have multiple orders. Orders contain multiple products. I must calculate average spend per customer per product category. Help me write optimized SQL that avoids duplicate counting and handles NULL values correctly. Explain your reasoning for each design decision."

Prompt:

"Optimize this Snowflake query for a fact table with 100M+ rows. Table is clustered by date and region. Query filters by date range and aggregates by product_id. Use QUALIFY clauses and result-set caching where appropriate. Rewrite the query to minimize compute cost and micro-partition scans."

Prompt:

"My SQL query is returning duplicate rows when joining orders and order_items. Each order can have multiple items. I only want one row per order with total quantity and total amount. How should I structure the query to avoid duplicates? Show the corrected SQL and explain why the original was duplicating."

Prompt:

"I have a customer DataFrame with these issues: Missing values in income and age columns (15% and 8% respectively), Duplicate customer records based on email, Inconsistent formatting in city names (mixed case, abbreviations), Outlier ages above 120 that are data entry errors. Generate pandas code to: Remove duplicate records keeping the most recent entry, Impute missing age with median and missing income with mean by customer_segment, Standardize city names to title case and expand common abbreviations, Cap age outliers at 100, Create a data quality report showing before/after statistics. Include explanatory comments for each step."

Prompt:

"I have an e-commerce dataset with columns: customer_id, order_date, product_category, order_value, payment_method, shipping_region. I want to investigate: Seasonal purchasing patterns across product categories, Products frequently purchased together, Customer segments based on purchasing behavior and value, Regional differences in product preferences. For each investigation: Suggest specific analytical approaches, Recommend appropriate visualizations, Identify relevant summary statistics, Highlight potential pitfalls or considerations. Prioritize insights with clear business implications."

Prompt:

"Given a customer dataset with columns: age, signup_date, last_purchase_date, region, total_spent, order_count, average_order_value, suggest 5 new features that could improve purchase prediction accuracy. For each feature: Provide the feature name and description, Explain the intuition for why it helps prediction, Write pandas code to create it, Note any assumptions or limitations."

Prompt:

"Create a bar chart for monthly sales data with: Month on the x-axis, revenue on the y-axis, Different colors for each product category, A secondary axis showing order count as a line, A title and clear legend. Output the code in a clean, well-commented format using seaborn or matplotlib."

Prompt:

"Summarize these model results for executives with no technical background: Model: Random Forest for customer churn prediction, Accuracy: 84%, Most important features: (1) days_since_last_order, (2) customer_tenure, (3) support_tickets_count, Business impact: Identifying high-risk customers 2 weeks earlier enables targeted retention campaigns. Create a 3–4 sentence summary highlighting business value and key findings. Avoid statistical jargon."

Prompt:

"Generate clear descriptions for these variables in a data dictionary: cust_ltv (numeric, range 0–50,000), churn_risk_score (numeric, range 0–1), engagement_level (categorical: low, medium, high), pref_channel (categorical: email, phone, chat, none). Follow this format: Variable name and type, Business definition in plain language, Valid values or range, Calculation method or source, Usage notes or caveats."

Prompt:

"I'm designing an A/B test for a SaaS product where we're testing a new checkout flow. Control group: 5,000 users, current checkout flow. Treatment group: 5,000 users, new checkout flow. Duration: 2 weeks. Primary metric: Conversion rate. Secondary metrics: Average order value, time to purchase. Review this experimental design and identify potential issues (e.g., statistical power, bias sources, business context). Suggest improvements to the design."

Prompt:

"I need to analyze customer churn patterns in subscription data. Before providing recommendations: First, clarify what key metrics and variables would be most relevant for churn analysis. Then, confirm which analytical approaches would be appropriate given these variables. Finally, provide a structured analysis plan including data preparation, modeling approach, and evaluation criteria. Walk through your reasoning at each step."

Prompt:

"I want to build a function detecting outliers in financial transaction data. Before writing code: Clarify what outlier detection methods suit financial data. Confirm the advantages and disadvantages of each approach and recommend one. Complete by writing a Python function implementing the recommended method. Address each step before moving to the next."