Skip to content

README.md

Latest commit

 

History

History
52 lines (33 loc) · 2.57 KB

README.md

File metadata and controls

52 lines (33 loc) · 2.57 KB
Raw

Python AI Debugging Workflow

AI-Powered Debugging and Refactoring

This repository contains my solution for the Datacom AI-Powered Debugging and Refactoring Simulation. It demonstrates a structured, professional approach to diagnosing and fixing a legacy Python data processing script (process_data.py) that was intermittently failing and missing service-level agreement (SLA) performance targets.

By leveraging AI tools as a technical co-pilot, this project highlights core competencies in maintaining and optimizing data pipelines, a critical skill for robust data analytics and engineering.

🔍 Project Objectives

  • Codebase Comprehension: Analyze and document the functionality of undocumented legacy Python code.
  • Root-Cause Analysis: Diagnose critical bugs by analyzing system logs and execution tracebacks.
  • Test-Driven Fixing: Develop a targeted, failing unit test (TEST_CASES.py) to reliably reproduce the issue before altering the code.
  • Performance Optimization: Refactor inefficient logic (e.g., O(N^2) nested loops) into highly performant solutions to meet strict SLAs.
  • Continuous Documentation: Maintain a comprehensive journal (DEBUG_LOG.md) detailing the AI prompts, reasoning, and step-by-step workflow.

📂 Repository Structure

datacom-debugging-simulation/
├── process_data.py           # The original legacy script containing the bug
├── refractored_function.py   # The isolated, optimized function logic
├── TEST_CASES.py             # Unit tests written to reproduce and validate the fix
├── DEBUG_LOG.md              # Detailed journal of the AI-assisted debugging process
├── README.md                 # Project overview and instructions
└── error.log                 # The sample production error log used for diagnosis


📝 Documentation
Transparency and tracking are critical when utilizing AI for code generation and refactoring.
All debugging notes, diagnostic reasoning,and specific prompts used during this task are
stored in a dedicated file:

DEBUG_LOG.md: A comprehensive, step-by-step journal detailing the AI-assisted debugging
and root-cause analysis workflow.

🛠️ Tools & Technologies Used
Language: Python 3.x

Testing: unittest framework

Development Environment: Visual Studio Code (VS Code)

AI Assistants: * GitHub Copilot Chat (for code context and refactoring)

Google Gemini (for prompt optimization and architectural reasoning)

Core Competencies: AI-Assisted Code Review, Test-Driven Development (TDD), and Pipeline Optimization

Context: Datacom Job Simulation on Forage - Task 1