Fijjy
A distributed podcast operations platform — recording workflows, AI-generated show notes and social content, WhisperX transcription, and queue-based multi-channel publishing — that a real 95-episode podcast now runs on.
Challenge
Running a podcast end to end means four or five separate tools stitched together by hand: a recording setup, a transcription service, a writer for show notes and social copy, and a publishing workflow per platform. Every new episode multiplied the manual work, and none of the pieces talked to each other.
Approach
We built Fijjy as a single operations platform instead of a bundle of subscriptions. A Go API, built contract-first with OpenAPI and backed by MongoDB, is the system of record for recordings, guests, and production state. An async AI pipeline (Python, FastAPI, Celery) generates show notes, summaries, and social snippets in parallel across worker queues. WhisperX handles transcription as its own asynchronous workflow, sized for long-form audio. Queue-based workers then fan a published episode out to X/Twitter, Facebook, and Instagram on event-driven triggers. Hosting runs on self-hosted Castopod, which Fijjy drives for RSS generation, so the show owns its feed instead of renting it.
Outcome
The proof point is public and unforgiving: Mentoring Developers, a podcast with 95+ published episodes, migrated its entire archive and live RSS feed from captivate.fm onto Fijjy's infrastructure. Every historical episode, its metadata, and the feed that podcast apps subscribe to survived the cutover — the kind of migration where an error means a real audience notices immediately. The show runs on Fijjy today.