TScrapingBot

TScrapingBot is a high-performance Telegram agent designed for automated web scraping and real-time data analysis. Originally conceived in Python, the project has been completely rewritten in Rust to maximize concurrency and resource efficiency.

Key Features:

• Ultra-Fast Inference Engine: LLM integration via Groq, delivering near-instant responses.
• Robust Architecture: Built with Teloxide for asynchronous, scalable bot management.
• Data & Persistence: Secure, efficient database handling through SQLx connected to Supabase (PostgreSQL).
• Intelligent Analysis: Advanced reasoning capabilities and context memory for complex queries on extracted web content.

I automated the bot’s lifecycle with CI/CD pipelines using GitHub Actions, incorporating unit testing and continuous deployment. Additionally, I implemented cross-compilation to generate optimized binaries for multiple architectures with each release.

Bringing this project to life has been a truly fascinating journey. The idea stemmed from my desire to find a way to analyze any webpage using AI, making it as simple as sending a message. During the process, I delved into the intricacies of Telegram updates and gained valuable experience with Supabase and Render deployments.

This project marked my first significant tool as a programmer—one I continue to use and refine to this day.