Checklist the per-video bar — engine/sim
84.5/100
quantitative quality · weights learn from your reviews (
engine.sim.memory review thread-build-log-W33-Wed good|bad)⚠ 6 flag(s) — not yet ship-ready: copy_genericlow_vo_coveragedead_airgeneric_scene · see docs/strategy/VIDEO-CHECKLIST.md
Composition comp · template family · expected output
9:16 Reelrendered1:1 Squarepending16:9 Widepending9:16 4Kpending1:1 4Kpending16:9 4KpendingGIF (SMS)pending
expected output: 1/7 rendered — same matrix the /media preview surfaces for this asset.
Composition layer × scene 6 scenes · 36.2s · comp_id + rendered still + tier + the script
| # | Layer (comp_id · still · tier) | Beat | Timecode | Motion | Logo | Audio | VO / on-screen / caption |
|---|---|---|---|---|---|---|---|
| 1 |  matches intent shared field signature-3d | open | 0–6.8s | spatial-parallax | icon·ember-fill | ♪ bed_in | Week eleven. We shipped automatic transcription this week, any audio or video to clean text locally. on-screen: Shipped: transcripts from any video |
expected on screen: black ground · knot hero in the shared Signal Field · Magister leads · node-graph · spatial-parallax · icon·ember-fill logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape knotground blacktreatment ember-fillmotion spatial-parallaxpower summoninstrument summon→Shipped: transcripts from any video | |||||||
| 2 |  matches intent JsonDiff template | build | 6.8–13.7s | type-on | icon·wireframe | ♪ node_lock | We used to pay per minute to transcribe interviews. Now a local speech model does it for free with word-level timing. on-screen: Old way: pay per minute to transcribe |
expected on screen: black ground · a JsonDiff panel over a dimmed Signal Field · Magister leads · code · type-on · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id JsonDiffvisual codeshape knotground blacktreatment wireframemotion type-onpower summoninstrument draw-on→Old way: pay per minute to transcribecurate codeLines, fileName, lines | |||||||
| 3 |  matches intent StackTrace template | diff | 13.7–19.799999999999997s | crossfade-8f | icon·wireframe | ♪ node_lock | It broke on sync. The timestamps drifted later and later, so captions lagged the speaker. on-screen: What broke: timestamps drifted late |
expected on screen: black ground · a StackTrace panel over a dimmed Signal Field · Magister leads · code · crossfade-8f · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id StackTracevisual codeshape knotground blacktreatment wireframemotion crossfade-8fpower morphinstrument morph→What broke: timestamps drifted latecurate codeLines, errMsg, errType, fix, frames | |||||||
| 4 |  matches intent SchematicCard template | step | 19.8–26.200000000000003s | kinetic-build | icon·wireframe | ♪ node_lock | Now any clip becomes searchable text in minutes. Next, we burn the captions straight onto the video. on-screen: Next: auto-burn the captions |
expected on screen: black ground · a SchematicCard panel over a dimmed Signal Field · Magister leads · pipeline · kinetic-build · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id SchematicCardvisual pipelineshape knotground blacktreatment wireframemotion kinetic-buildpower throwinstrument laser-trace→Next: auto-burn the captionscurate nodes, stages, steps | |||||||
| 5 |  first render · fix pending KpiGrid templatedead_airgeneric_scene | proof | 26.2–31.2s | receipts-count | icon·wireframe | ♪ node_lock | Build Log |
expected on screen: black ground · a KpiGrid panel over a dimmed Signal Field · Magister leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left spec (the prompt): comp_id KpiGridvisual receiptsshape knotground blacktreatment wireframemotion receipts-countpower receiptsinstrument spotlight→Build Logcurate kpis, statsLabels | |||||||
| 6 |  first render · fix pending shared field signature-3ddead_airgeneric_scene | resolve | 31.2–36.2s | coalescence | icon·ember-fill | ♪ bed_out | Build Log |
expected on screen: black ground · knot hero in the shared Signal Field · Magister leads · coalescence · coalescence · icon·ember-fill logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape knotground blacktreatment ember-fillmotion coalescencepower coalescenceinstrument coalescence→Build Log | |||||||
Format stack 3 aspects · same scenes[], re-cropped
9:16
1080×1920
Stories · TikTok · YouTube Shorts · Reels
1:1
1080×1080
LinkedIn · Facebook · Instagram
16:9
1920×1080
X/Twitter · YouTube · LinkedIn video
Channels 9 destinations
LinkedInX/TwitterYouTubeInstagramFacebookThreadsTikTokPinterestBluesky
Social captions supplemental published copy · per channel (comp_id level)
tiktokBuild log, week 11: we shipped automatic transcription. Any audio or video to clean text, locally, free, with word-level timing. It broke on sync at first, timestamps drifted late. Next we burn the captions onto the clip. (AI-assisted)
instagramBuild log: week 11.
Shipped: transcripts from any video.
Local. Free. Word-level timing.
What broke: timestamps drifted late.
Next: auto-burn the captions.
#buildlog #whisper #transcription #ai #captions
linkedinBuild log, week 11.
Shipped: automatic transcription. Any audio or video to clean text, running locally, with word-level timing. We used to pay per minute to transcribe interviews; now a local speech model does it for free.
What broke: sync. The timestamps drifted later and later, so captions lagged the speaker. Fixed.
Now any clip becomes searchable text in minutes. Next, we burn the captions straight onto the video.
xBuild log wk11: shipped local transcription. Any video to clean text, free, word-level timing. It broke on sync, timestamps drifted late. Next: burn captions onto the clip. temerarii.com/work
facebookBuild log, week 11. We shipped automatic transcription: any audio or video to clean text, running locally and free, with word-level timing. It broke on sync at first, the timestamps drifted late. Next we burn the captions straight onto the video. Follow along at temerarii.com/work.
threadsBuild log, week 11.
Shipped: automatic transcription. Any audio or video to clean text, locally, free, with word-level timing.
What broke: sync. Timestamps drifted later and later, so captions lagged the speaker. Fixed it.
Next: burn the captions straight onto the video.
pinterestBuild log: shipping a local automatic transcription tool. Turn any audio or video into clean text for free with word-level timing using a local speech model. The bug we hit was sync drift, where timestamps fell behind the speaker. Next step: burning captions onto the video.
blueskyBuild log wk11: shipped local transcription. Any video to clean text, free, word-level timing. It broke on sync, timestamps drifted late. Next: burn captions onto the clip.
youtubeTitle: Build Log Week 11 - Local Transcription, Free
This week we shipped automatic transcription: any audio or video to clean text, running locally and free, with word-level timing. We walk what broke (sync drift, where timestamps fell behind the speaker) and how we fixed it. Next up: burning captions straight onto the video.