Stories, craft notes, and engineering deep-dives from the team.
The build log of a tool that turns a prompt into a finished, narrated, on-brand video. We write about the things that actually moved the needle - voice markup, image consistency, render pipelines, honest pricing.
POV vs. omniscient narration: choosing the right narrator
The narrator you pick decides the video before the first frame is generated. A working framework for choosing POV, omniscient, or interviewer - with three worked examples.
Style anchoring: how to keep 40 AI images consistent
A reference frame, a style suffix, and a discipline. The protocol we use to hold a single visual language across forty Gemini frames in a five-minute video.
The 3-pass ffmpeg pipeline behind every VideoCue render
A walk through the render container: image-to-clip with Ken Burns motion, concat with crossfades, overlay-plus-audio-plus-watermark. And why we shell out to Python from Laravel.
Pricing AI video honestly: what each minute costs
A line-item cost breakdown of a 5-minute AI video - voice, images, music, render - and an honest comparison to a freelancer, Runway, and a stock subscription.
From script to upload in under an hour: a creator's workflow
A minute-by-minute walkthrough of producing a 5-minute narrative video, end to end, in 59 minutes - using a single creator, one topic, and a stopwatch.
One post in your inbox, monthly.
New craft notes, render deep-dives, and the occasional teardown of a video we shipped. No spam, ever. Unsubscribe in one click.