• Built an end-to-end content operations service for multi-brand pipelines with scheduled scraping and automated
media processing.
• Scraped YouTube ytInitialData using Cheerio and stored normalized videos, shorts, and episodes in MongoDB with
deduplication logic.
• Automatedamediaprocessing pipeline using yt-dlp and FFmpeg for H.264/AAC transcoding, poster extraction, and
platform uploads.
• Containerized the application with Docker including FFmpeg and yt-dlp, and integrated Jenkins for CI/CD readiness.