Bug Deduplication with AI (Node.js + OpenRouter)

QA teams often face the same problem: duplicate bug reports. Different test cases fail, but the root cause is actually the same. To cut through the noise, I built a Proof of Concept (PoC) using Node.js + OpenRouter AI to automatically cluster similar failures and suggest root causes

Why?

🚫 Less time wasted triaging duplicate bugs
🔍 Faster root cause analysis
🤖 AI-driven clustering & summarization

Tech Stack

• Node.js
• ml-kmeans for clustering
• OpenRouter API (chat model for semantic grouping)

Setup
Install dependencies:
npm install node-fetch ml-kmeans dotenv

Add your API key to .env:
OPENROUTER_API_KEY=your_api_key_here
CHAT_MODEL=google/gemma-3n-e2b-it:free


Example snippet:

Example result:

Takeaways

✅ AI reduces noise from duplicate bugs
✅ Saves QA team time during triage
✅ Extensible to native CI/CD logs

This is just a PoC, but it shows how AI can make QA workflows smarter. What do you think of this POC? If you have any suggestions, leave a comment

Or if you want to read the full version, you can click here