Closing the Loop: Executing LLM-Generated SQL and Presenting Results

2026-04-23 · Source: Naturallanguageprocessing on Medium · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Data Science & Analytics · Depth: Intermediate, medium

Summary

This article, "Closing the Loop: Executing LLM-Generated SQL and Presenting Results," is the third part of a series detailing a production-ready pipeline for converting natural language questions into executable SQL queries. It introduces two new Python modules: `SQLExecutor` and `ResultFormatter`. The `SQLExecutor` uses the `pyodbc` library to connect to MS SQL Server, execute validated SQL strings, and return structured `ExecutionResult` objects, ensuring no unhandled exceptions. The `ResultFormatter` then takes these results and formats them into a human-readable terminal table or exports them to CSV, without external dependencies. The article demonstrates the full pipeline, from schema extraction and LLM-based SQL generation to validation, execution, and formatted output, handling valid queries, blocked dangerous operations, and empty result sets gracefully.

Key takeaway

For AI Engineers building LLM-to-SQL applications, integrating robust execution and formatting modules is crucial for production readiness. You should implement a dedicated `SQLExecutor` that handles database interactions and errors gracefully, alongside a `ResultFormatter` for clear output. Additionally, consider adding post-validation checks for hallucinated column names and result set size limits to enhance system reliability and prevent unexpected issues in live environments.

Key insights

A robust LLM-to-SQL pipeline requires secure execution and clear result presentation.

Principles

• Modular design enhances flexibility.
• Defense in depth prevents SQL injection.
• Structured error handling is critical.

Method

The proposed method involves a pipeline: SchemaExtractor → SimpleSQLQueryGenerator → SQLValidator → SQLExecutor → ResultFormatter, ensuring schema-awareness, safety, execution, and presentation.

In practice

• Use `pyodbc` for MS SQL Server connections.
• Implement `autocommit=False` for safety.
• Return structured `ExecutionResult` objects.

Topics

• LLM-Generated SQL
• SQL Query Execution
• pyodbc Library
• Result Formatting
• SQL Validation

Code references

• sainathudata/simple_sql_query_generator_llm

Best for: AI Engineer, Machine Learning Engineer, Software Engineer

Related on AIssential

• Teaching Your SQL Generator to Fix Its Own Mistakes — LLMs can self-correct runtime SQL errors by processing database feedback in a retry loop.
• Testing SQL Like a Software Engineer: Unit Testing, CI/CD, and Data Quality Automation — Treat SQL like software by implementing version control, unit tests, and automated data quality checks.
• AI Builders: Building a text-to-SQL agent — Text-to-SQL agents democratize data access by translating natural language queries into structured database commands.
• LLM Eval Workflow: How to Build Reliable AI Quality Gates Without Vibes — Reliable LLM evaluation requires a structured workflow integrating diverse checks and continuous feedback, not just tools.
• Turing Award Winner: Postgres, Disagreeing with Google, Future Problems | Mike Stonebraker — Specialized database architectures and strong data integrity are crucial for performance and correctness, especially as applications become complex.
• Deterministic AI at Work: The LLM Is the Most Unreliable Function in My Pipeline, So I Treat It… — Treat LLMs as unreliable functions within a deterministic code pipeline to achieve reliable, auditable system outputs.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Naturallanguageprocessing on Medium.