Build a Data Pipeline — SQL + Python + AI

pip install anthropic pandas

# Mac / Linux
export ANTHROPIC_API_KEY="sk-ant-..."

# Windows (Command Prompt)
set ANTHROPIC_API_KEY=sk-ant-...

import sqlite3
import pandas as pd

# Connect to the database
conn = sqlite3.connect("sales.db")

# Run a SQL query — get results as a DataFrame
query = """
SELECT c.name AS customer, c.segment, COUNT(*) AS orders, SUM(o.amount) AS total_revenue, AVG(o.amount) AS avg_order
FROM orders o
JOIN customers c ON o.customer_id = c.id
WHERE o.status = 'completed'
GROUP BY c.name, c.segment
ORDER BY total_revenue DESC
"""

df = pd.read_sql_query(query, conn)
conn.close()

print(df.to_string(index=False))

import sqlite3
import pandas as pd
import anthropic
import os

# 1. Query the database
conn = sqlite3.connect("sales.db")
query = """
SELECT c.name, c.segment, COUNT(*) AS orders, ROUND(SUM(o.amount), 2) AS total_revenue, ROUND(AVG(o.amount), 2) AS avg_order
FROM orders o
JOIN customers c ON o.customer_id = c.id
WHERE o.status = 'completed'
GROUP BY c.name, c.segment
ORDER BY total_revenue DESC
"""
df = pd.read_sql_query(query, conn)
conn.close()

# 2. Format as a clean text table
data_text = df.to_string(index=False)
total = df['total_revenue'].sum()

# 3. Send to Claude for analysis
client = anthropic.Anthropic()

prompt = f"""Here is a sales revenue summary from our database:

{data_text}

Total revenue: ${total:,.2f}

Please analyze this data in 3-4 sentences. Identify:
- Who the top customer is and their share of revenue
- Any patterns in the segment data
- One actionable recommendation for the sales team"""

message = client.messages.create( model="claude-3-5-haiku-20241022", max_tokens=512, messages=[{"role": "user", "content": prompt}]
)

# 4. Print the analysis
print("\n=== DATA ===\n")
print(data_text)
print(f"\nTotal revenue: ${total:,.2f}")
print("\n=== AI ANALYSIS ===\n")
print(message.content[0].text)

def run_sales_report(db_path: str, query: str, prompt_template: str) -> str: """Query a SQLite DB, format results, get AI analysis. Returns the analysis.""" import sqlite3, pandas as pd, anthropic # Query conn = sqlite3.connect(db_path) df = pd.read_sql_query(query, conn) conn.close() # Analyze client = anthropic.Anthropic() prompt = prompt_template.format(data=df.to_string(index=False)) msg = client.messages.create( model="claude-3-5-haiku-20241022", max_tokens=512, messages=[{"role": "user", "content": prompt}] ) return msg.content[0].text

# Use it
if __name__ == "__main__": analysis = run_sales_report( db_path="sales.db", query="SELECT name, SUM(amount) AS total FROM orders o JOIN customers c ON o.customer_id=c.id GROUP BY name ORDER BY total DESC", prompt_template="Analyze this sales data in 3 sentences:\n\n{data}" ) print(analysis)