Leviro Jobs Hunter

Leviro Jobs Hunter is a Leviro service for freelancers who want relevant job listings in Telegram. It scans freelance marketplaces, matches listings to a subscriber's interests, and can draft a proposal for a selected job. The product grew out of earlier internal job monitors.

The problem

Freelancers otherwise check several marketplaces repeatedly and sort through listings that do not fit their skills or budget. A useful alert needs a fresh listing, a clear match, and delivery that respects each subscriber's settings. The bot keeps those steps together.

Key features

• Scrapers collect listings from Khamsat, Mostaql, Nafezly, Kafiil, and Freelancer.com.
• Skill and category matching filters jobs before personal delivery.
• Subscribers can change platform, keyword, client blocklist, and budget filters in Telegram.
• Private job alerts use delivery queues; public channels receive selected listings.
• Quiet hours and digest settings control when notifications arrive.
• The bot generates and revises proposals using a freelancer profile and job details.
• Telegram Stars handles subscription purchases and AI credit top-ups.
• An admin dashboard exposes scraper controls, users, subscriptions, proposals, and logs.

Technical architecture

A Node.js backend runs scrapers, the Telegram bot, delivery workers, and an internal admin API. Scrapers publish new-job events through Redis. The dispatcher reads stored jobs and subscriber preferences from MongoDB, applies matching rules, and adds delivery work to BullMQ. Sender workers call Telegram. The React dashboard talks to the internal API; a separate static site serves English and Arabic product pages. Gemini generates proposals, with another Gemini model and Groq available as fallback providers.

Engineering highlights

• Redis rate limiting and queue controls pace Telegram delivery across workers.
• Admin access uses authenticated sessions stored in Redis and a protected API.
• Proposal generation validates provider output and retries eligible failures.
• Scraping includes URL checks and a browser path for sites that challenge plain HTTP requests.