Checklist the per-video bar — engine/sim
81.0/100
quantitative quality · weights learn from your reviews (
engine.sim.memory review thread-minus-the-hype-W33-Mon 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 5 scenes · 26.9s · 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 | hook | 0–5.0s | kinetic-build | icon·liquid-chrome | ♪ node_lock | Drop in a thousand PDFs and ask anything, they promise, and the demo always works on a clean slide deck. on-screen: "AI reads every PDF" |
expected on screen: red ground · tetra hero in the shared Signal Field · Nexus leads · mark · kinetic-build · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual markshape tetraground redtreatment liquid-chromemotion kinetic-buildpower laser-lockinstrument laser-trace→"AI reads every PDF" | |||||||
| 2 |  matches intent BuildLog template | diff | 5.0–10.8s | crossfade-8f | icon·white-knockout | ♪ node_lock | What works: chunk the docs, embed them, and force the model to quote the chunk it used. on-screen: Chunk, embed, cite |
expected on screen: red ground · a BuildLog panel over a dimmed Signal Field · Nexus leads · code · crossfade-8f · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id BuildLogvisual codeshape tetraground redtreatment white-knockoutmotion crossfade-8fpower morphinstrument morph→Chunk, embed, citecurate codeLines, lines | |||||||
| 3 |  matches intent ChecklistCard template | teach | 10.8–16.9s | kinetic-build | icon·liquid-chrome | ♪ node_lock | The cost is building and refreshing the index. Pick your ten most-asked docs and embed those first. on-screen: The cost: an index |
expected on screen: red ground · a ChecklistCard panel over a dimmed Signal Field · Nexus leads · node-graph · kinetic-build · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id ChecklistCardvisual node-graphshape tetraground redtreatment liquid-chromemotion kinetic-buildpower morphinstrument morph+laser→The cost: an indexcurate items, nodes | |||||||
| 4 |  first render · fix pending ListCard templatedead_airgeneric_scene | proof | 16.9–21.9s | receipts-count | icon·white-knockout | ♪ node_lock | Minus the Hype |
expected on screen: red ground · a ListCard panel over a dimmed Signal Field · Nexus leads · receipts · receipts-count · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id ListCardvisual receiptsshape tetraground redtreatment white-knockoutmotion receipts-countpower receiptsinstrument spotlight→Minus the Hypecurate items, statsLabels | |||||||
| 5 |  first render · fix pending shared field signature-3ddead_airgeneric_scene | resolve | 21.9–26.9s | coalescence | icon·white-knockout | ♪ bed_out | Minus the Hype |
expected on screen: red ground · tetra hero in the shared Signal Field · Nexus leads · coalescence · coalescence · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape tetraground redtreatment white-knockoutmotion coalescencepower coalescenceinstrument coalescence→Minus the Hype | |||||||
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)
tiktokThey say drop in a thousand PDFs and ask anything. What works: chunk the docs, embed them, and force the model to quote the chunk it used. The cost is building the index. Start with your 10 most-asked docs. (AI-assisted)
instagram"AI reads every PDF."
What works:
Chunk the docs.
Embed them.
Force the model to quote the chunk it used.
The cost is building and refreshing the index. Start with your 10 most-asked docs.
#ai #rag #embeddings #documents #search
linkedinThe pitch: drop in a thousand PDFs and ask anything.
What actually works: chunk the docs, embed them, and force the model to quote the chunk it used. That citation step is what keeps it honest.
The cost nobody mentions is building and refreshing the index. So don't boil the ocean. Pick your ten most-asked documents and embed those first. Get one answer with a real citation before you scale.
x"Drop in 1,000 PDFs, ask anything." What works: chunk the docs, embed them, force the model to quote the chunk it used. The cost is the index. Start with your 10 most-asked docs.
facebookThey promise you can drop in a thousand PDFs and ask anything. What works: chunk the docs, embed them, and force the model to quote the chunk it used. The cost is building and refreshing the index, so start with your ten most-asked docs and embed those first.
threads"Drop in a thousand PDFs and ask anything."
What works: chunk the docs, embed them, and force the model to quote the chunk it used so you can check it. The real cost is building and refreshing the index. Pick your 10 most-asked docs and embed those first.
pinterestHow to make AI actually answer from your PDFs: chunk the documents, embed them into a vector index, and force the model to quote the chunk it used. The hidden cost is building and refreshing that index, so start with your ten most-asked documents first.
bluesky"Drop in 1,000 PDFs, ask anything." What works: chunk the docs, embed them, force the model to quote the chunk it used. The cost is the index. Start with your 10 most-asked docs.
youtubeTitle: AI Won't Read Every PDF (here's what works)
The pitch is drop in a thousand PDFs and ask anything. The method that works: chunk the docs, embed them, and force the model to quote the chunk it used so you can verify it. The cost is building and refreshing the index, so start with your ten most-asked documents first.