HomerTrakker
An end-to-end automated Python pipeline that polls MLB’s StatsAPI in real time for home runs, downloads broadcast highlights, compiles vertical Shorts via FFmpeg, and auto-uploads to YouTube with Statcast-enriched metadata. 700+ videos published, 500K+ views on @SportsTrakker.
Background / Architecture
What this system proves
Problem
MLB highlights are released across multiple sources, making it time-consuming for content teams to aggregate and post clips in real time. Teams waste hours manually tracking, matching, and editing videos.
Solution
Built an end-to-end automated pipeline using n8n and custom JavaScript that: • Polls MLB’s StatsAPI in real time for home runs • Matches each event to its official highlight video • Generates ready-to-post short-form clips with auto-generated captions • Handles timezone differences across the full MLB slate and can be extended to other sports.
Impact
Processes every MLB game day automatically, eliminating manual editing. Saves content teams hours per game day while ensuring accuracy and timeliness. Demonstrates scalable workflow automation that combines product, ops, and technical problem-solving.