Publishing · Literary content

Project Marginalia — 400+ literary shorts from one input file.

Engagement: 2024 Q1 → maintenance
Brand confidential · engagement detail available under NDA
Brief

A literary content brand had a backlog of curated quotes and a posting goal that no human production timeline could meet. The brief: take one markdown file in — ship a year’s worth of upload-ready vertical videos out, with no creative work in between.

SectorPublishing · literary / educational content
Stage at startCurated 409-quote backlog. No creators, no audio, no covers, no video.
Channels in scopeYouTube Shorts (primary). Optional repurposing for Reels and TikTok.
ConstraintsSingle-operator program. Minimal per-short cost. Voice quality non-negotiable.
Build

01 · Parse

Single markdown file parsed into 409 atomic units — title, author, body. Idempotent, restartable, indexed.

Range flags allow batch generation, dry-runs, and audio-only reruns without re-rendering video.

02 · Compose

ElevenLabs voice synthesis per quote. Cover image lookup with auto-generated dark placeholder fallback. Vertical 1080×1920 composition with quote typography, attribution, and pan / zoom motion.

Output is upload-ready MP4 with mixed audio at consistent loudness.

03 · Ship

Shorts numbered sequentially and queued for upload. Brand can drag-drop or hand off to scheduled uploader.

Total runtime for the full backlog measured in hours, not weeks.

Result
409
finished shorts produced from one input file
< $1
marginal cost per finished short, all-in
0
manual editing or production hours
1080×1920
vertical, captioned, voiced, branded
~1 yr
of daily Shorts inventory from a single run
N+1
scaling — new quotes append, pipeline re-runs
Why this case matters

The point of Marginalia is not the literary niche — it is the cost structure. Once the pipeline exists, the unit economics of branded short-form content collapse from agency-priced ($500–$2,000 per short) to API-priced (under a dollar). The same shape works for any brand whose content has a curated input and a daily output expectation: book publishers, course creators, newsletter brands, religious orgs, B2B thought leadership programs, even internal-comms teams at large companies. The engine is the asset; the input is interchangeable.

Stack
ElevenLabsMoviePyPillowFFmpegPython 3.10Markdown parsingYouTube Data API
Run this for your library →

If you have a content backlog and a daily output goal that doesn’t pencil out manually, this engagement template applies directly.