Case Study
Telegram assistant that surfaces freelance gigs in real time
A scraping and automation bot that monitors Upwork job listings and delivers filtered gig alerts through Telegram. Built to reduce search overhead and surface relevant opportunities shortly after posting.
Manual Upwork searching was slow and inconsistent, making it easy to miss high-fit jobs early. There was a need for a reliable assistant that could detect and forward relevant gigs without constant dashboard polling.
I built a Node.js + Puppeteer bot with keyword filtering, deduplication, and Telegram delivery. The bot uses scheduled scraping intervals, resilient retry logic, and MongoDB persistence to maintain state across runs.
Telegram
Delivery Channel
Instant push notifications
MongoDB
Persistence
Deduplication and state tracking
Scheduled
Runtime
Predictable scan intervals
The bot uses Puppeteer for listing extraction, MongoDB for persistent deduplication, and Telegram Bot API for delivery. A scheduler controls scan cadence while retry + backoff logic handles transient scraping failures.
Challenge
Preventing noisy or duplicate alerts
Solution
Added persistent fingerprinting in MongoDB and structured keyword filters so only useful, new opportunities were delivered.
Challenge
Maintaining reliable scraping over repeated runs
Solution
Implemented retry/backoff and defensive DOM extraction patterns to reduce failure frequency during layout shifts.
UNFCCC - Automation Engineer. Chrome extension with Node.js and Puppeteer that replaced manual stream monitoring with automated quality checks and fast alerts.
Web Development case study using React, Tailwind CSS, Styled-Components, Node.js, and Express for clean, quick data-management operations.