
Webhooks for long-running video jobs: reliable pipeline patterns
Webhooks for long running video jobs fail quietly. Use the hint-plus-reconciler pattern: verify signatures, key idempotency on job ID, add a polling backstop.
73 articles about automation.

Webhooks for long running video jobs fail quietly. Use the hint-plus-reconciler pattern: verify signatures, key idempotency on job ID, add a polling backstop.

The extract subtitles FFmpeg one-liner is easy. Finding the right track with ffprobe, handling PGS bitmaps, and fixing mov_text in MP4 is where jobs break.

ffmpeg cropdetect finds black bars on one file fine, then breaks in a batch. Sample past the intro, take the last crop line, round to even, crop per file.

An AI dubbing workflow returns a translated track longer than the video. Measure both with ffprobe, fix the drift with atempo or apad, then re-mux cleanly.

Which n8n FFmpeg node should you use? The real split is local binary vs hosted API. Compare the community nodes, what breaks each, and what runs on n8n Cloud.

The ffmpeg drawvg filter draws animated lower thirds and progress bars from a script, with no pre-rendered PNGs. Working VGS code plus the 8.1 build catch.
AI avatar video automation stalls after the talking head renders. The assembly chain: stitch b-roll, burn captions, add a music bed, and ship 9:16 vertical.

The AI video generator workflow nobody documents: extract last frames, normalize mismatched clips, concat, lay an audio bed, burn captions once with FFmpeg.

Convert MOV to MP4 automatically with one API call. iPhone HEVC footage breaks web pipelines. Catch uploads in n8n, Make, or Zapier and ship MP4.

n8n large video files crash the instance because binary data sits in memory. Learn the real cause, the env vars that buy headroom, and the URL-passing fix.