Back Office · office.temerarii.xyz
One asset, all the way in — composition, the wireframe + storyboard, the output format stack, and the template, all read from the SAME content-index record. The expected output matches what /media surfaces for this post.
post social-W30-Sun-1kind threadweek W30date 2026-07-26campaign thread · Sunpillar multimediabeat Sunasset videoduration 53.0sground blackscenes 9

Checklist the per-video bar — engine/sim

88.1/100
plain languagevo coverageno dead airuniquenesscaption fitcompletenesscleanliness
quantitative quality · weights learn from your reviews (engine.sim.memory review social-W30-Sun-1 good|bad)
⚠ 2 flag(s) — not yet ship-ready: dup_sequence_across_assetsdup_set_across_assets · see docs/strategy/VIDEO-CHECKLIST.md

Composition comp · template family · expected output

composition SceneReelfamily / template SignalFieldReel
9:16 Reelpending1:1 Squarepending16:9 Widepending9:16 4Kpending1:1 4Kpending16:9 4KpendingGIF (SMS)pending
render pending — silent master not yet on disk
expected output: 0/7 rendered — same matrix the /media preview surfaces for this asset.

Composition layer × scene 9 scenes · 53.0s · comp_id + rendered still + tier + the script

#Layer (comp_id · still · tier)BeatTimecodeMotionLogoAudioVO / on-screen / caption
1s1
matches intent
shared field
signature-3d		open0–5.5sspatial-parallaxicon·ember-fill♪ bed_in
An interview show used to eat a whole crew and a week of edits.
on-screen: The interview used to eat days
expected on screen: black ground · octa hero in the shared Signal Field · Lumen leads · node-graph · spatial-parallax · icon·ember-fill logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape octaground blacktreatment ember-fillmotion spatial-parallaxpower summoninstrument summon→The interview used to eat days
2s2
matches intent
shared field
signature-3d		hook5.5–11.4skinetic-buildicon·wireframe♪ node_lock
Now the talk almost edits itself, and the guest never has to repeat a thing.
on-screen: Now the talk edits itself
expected on screen: black ground · octa hero in the shared Signal Field · Lumen leads · mark · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual markshape octaground blacktreatment wireframemotion kinetic-buildpower laser-lockinstrument laser-trace→Now the talk edits itself
3s3
matches intent
NumberedList
template		teach11.4–16.6skinetic-buildicon·wireframe♪ node_lock
Record the raw chat, then run the audio through Whisper for word-level timing.
on-screen: Record raw, then transcribe with Whisper
expected on screen: black ground · a NumberedList panel over a dimmed Signal Field · Lumen leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id NumberedListvisual node-graphshape octaground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Record raw, then transcribe with Whispercurate items, nodes
4s4
matches intent
ComparisonTable
template		proof16.6–22.1sreceipts-counticon·wireframe♪ node_lock
Those timestamps let The Big T-M jump straight to each clean answer, no scrubbing.
on-screen: Timestamps find every clean answer
expected on screen: black ground · a ComparisonTable panel over a dimmed Signal Field · Lumen leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ComparisonTablevisual receiptsshape octaground blacktreatment wireframemotion receipts-countpower receiptsinstrument spotlight→Timestamps find every clean answercurate colA, colB, rows, statsLabels
5s5
matches intent
JsonDiff
template		diff22.1–28.700000000000003scrossfade-8ficon·wireframe♪ node_lock
Most tools cut on silence. We cut on meaning, so the point lands and the filler dies.
on-screen: We cut on meaning, not silence
expected on screen: black ground · a JsonDiff panel over a dimmed Signal Field · Lumen leads · code · crossfade-8f · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id JsonDiffvisual codeshape octaground blacktreatment wireframemotion crossfade-8fpower morphinstrument morph→We cut on meaning, not silencecurate codeLines, fileName, lines
6s6
matches intent
ChecklistCard
template		teach28.7–35.3skinetic-buildicon·wireframe♪ node_lock
Feed the transcript to a model and ask it for the strongest forty seconds for each topic.
on-screen: Ask the model for the best 40
expected on screen: black ground · a ChecklistCard panel over a dimmed Signal Field · Lumen leads · node-graph · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id ChecklistCardvisual node-graphshape octaground blacktreatment wireframemotion kinetic-buildpower morphinstrument morph+laser→Ask the model for the best 40 secondscurate items, nodes
7s7
matches intent
KpiGrid
template		proof35.3–40.8sreceipts-counticon·wireframe♪ node_lock
One sit-down becomes ten honest clips, each one tagged with the exact words spoken.
on-screen: One sit-down becomes ten clips
expected on screen: black ground · a KpiGrid panel over a dimmed Signal Field · Lumen leads · receipts · receipts-count · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id KpiGridvisual receiptsshape octaground blacktreatment wireframemotion receipts-countpower receiptsinstrument spotlight→One sit-down becomes ten clipscurate kpis, statsLabels
8s8
matches intent
FlowSchematic
template		step40.8–46.699999999999996skinetic-buildicon·wireframe♪ node_lock
Your move: transcribe first, pick by timestamp, then trim. The edit gets boring and fast.
on-screen: Your move: timestamp before you trim
expected on screen: black ground · a FlowSchematic panel over a dimmed Signal Field · Lumen leads · pipeline · kinetic-build · icon·wireframe logo · caption bottom-left
spec (the prompt): comp_id FlowSchematicvisual pipelineshape octaground blacktreatment wireframemotion kinetic-buildpower throwinstrument laser-trace→Your move: timestamp before you trimcurate stages, steps
9s9
matches intent
shared field
signature-3d		resolve46.7–53.0scoalescenceicon·ember-fill♪ bed_out
Talk once, publish all week. The conversation stops being a project and starts being a habit.
on-screen: Talk once, publish all week
expected on screen: black ground · octa hero in the shared Signal Field · Lumen leads · coalescence · coalescence · icon·ember-fill logo · caption bottom-left
spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape octaground blacktreatment ember-fillmotion coalescencepower coalescenceinstrument coalescence→Talk once, publish all week

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)

tiktokPOV: you stop scrubbing interview footage forever. Transcribe with Whisper, get word-level timestamps, then cut on meaning instead of silence. One sit-down = ten clips. (AI-assisted edit, real human guest.)
instagramOne conversation. Ten clips. Transcribe first → cut on meaning → publish all week. The talk edits itself. #interviews #contentworkflow #aiediting #creatorlife #temerarii
linkedinInterview content used to mean a crew and a week of edits. The unlock isn't a fancy camera, it's a transcript. Run the audio through Whisper for word-level timestamps, jump straight to each clean answer, and cut on meaning instead of silence. One sit-down becomes ten clips, each tagged with the exact words spoken. Takeaway: transcribe before you trim.
xStop scrubbing interview footage. Transcribe with Whisper, get word-level timestamps, cut on meaning not silence. One talk = ten clips. temerarii.xyz
facebookInterview shows used to eat a crew and a full week of edits. The fix is simple: transcribe the raw audio with Whisper, use the word-level timestamps to find every clean answer, and cut on meaning instead of silence. One sit-down turns into ten clips. Want the full workflow? It's on the site.
threadsInterview editing trick: transcribe the raw audio first, get word-level timestamps, then cut on meaning, not silence. One sit-down becomes ten clips. The talk basically edits itself.
pinterestHow to turn one interview into ten clips: transcribe audio with Whisper for word-level timestamps, find clean answers fast, and cut on meaning instead of silence. Interview editing workflow, content repurposing, video production tips for creators and small studios.
blueskyInterview editing: transcribe with Whisper → word-level timestamps → cut on meaning, not silence. One talk = ten clips. temerarii.xyz
youtubeHow One Interview Becomes Ten Clips (Transcribe-First Editing) We used to spend a week editing a single interview. Now the talk almost edits itself. In this short we break down the transcribe-first method: record the raw conversation, run the audio through Whisper for word-level timestamps, jump to each clean answer, and cut on meaning instead of silence. One sit-down becomes ten honest clips. Transcribe first, pick by timestamp, then trim. Made with The Big T-M. AI-assisted edit, real human guest. more at temerarii.xyz

Cross-links every lens is a view on this one record

→ Wireframe · scenes→ Calendar week→ Media (audience preview)→ Storyboard hub→ Day view→ Channel view