dev-ops-and-chill

📝 Postmortem: Twitter Scraping Woes (2025-04-02)

🧵 Problem

We needed to extract Trump’s latest tweet text every time the GitHub Actions workflow ran. Initially, we assumed this would be a simple scrape using a headless browser or public proxy.

🔥 What Went Wrong

🛠️ Attempted Fixes

✅ Final Solution

We built a standalone Node.js app using:

This modular separation allowed us to run the Node scraper independently from the C# haiku pipeline while avoiding all authentication, rate limiting, and detection headaches.

🤯 Lessons Learned

😎 Current Vibe

“We broke the internet to write poems.” — The Department of Vibes and Haiku Warfare