Checklist the per-video bar — engine/sim
98.5/100
quantitative quality · weights learn from your reviews (
engine.sim.memory review social-W27-Mon-1 good|bad)⚠ 1 flag(s) — not yet ship-ready: copy_generic · see docs/strategy/VIDEO-CHECKLIST.md
Composition comp · template family · expected output
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 · 66.2s · 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 | 0–7.3s | spatial-parallax | icon·liquid-chrome | ♪ bed_in | A few years back, using AI meant a research lab, a cluster of cards, and a very patient budget. on-screen: AI used to mean a research lab |
expected on screen: red ground · cone hero in the shared Signal Field · Augur leads · node-graph · spatial-parallax · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape coneground redtreatment liquid-chromemotion spatial-parallaxpower summoninstrument summon→AI used to mean a research lab | |||||||
| 2 |  matches intent shared field signature-3d | hook | 7.3–14.6s | kinetic-build | icon·white-knockout | ♪ node_lock | Now the same power is one API call. The hard part moved from building it to using it well. on-screen: Now it's an API call |
expected on screen: red ground · cone hero in the shared Signal Field · Augur leads · mark · kinetic-build · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual markshape coneground redtreatment white-knockoutmotion kinetic-buildpower laser-lockinstrument laser-trace→Now it's an API call | |||||||
| 3 |  matches intent NumberedList template | teach | 14.6–21.6s | kinetic-build | icon·liquid-chrome | ♪ node_lock | Stop pasting prompts into a chat box. Call the model from your code so the work repeats itself. on-screen: Stop chatting, start calling |
expected on screen: red ground · a NumberedList panel over a dimmed Signal Field · Augur leads · node-graph · kinetic-build · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id NumberedListvisual node-graphshape coneground redtreatment liquid-chromemotion kinetic-buildpower morphinstrument morph+laser→Stop chatting, start callingcurate items, nodes | |||||||
| 4 |  matches intent RankList template | proof | 21.6–28.900000000000002s | receipts-count | icon·white-knockout | ♪ node_lock | Ask for the answer as strict JSON, and now the model plugs straight into a spreadsheet or a database. on-screen: Force it to answer in JSON |
expected on screen: red ground · a RankList panel over a dimmed Signal Field · Augur leads · receipts · receipts-count · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id RankListvisual receiptsshape coneground redtreatment white-knockoutmotion receipts-countpower receiptsinstrument spotlight→Force it to answer in JSONcurate rows, statsLabels | |||||||
| 5 |  matches intent JsonDiff template | diff | 28.9–37.0s | crossfade-8f | icon·liquid-chrome | ♪ node_lock | The win is not a smarter chatbot. It is a function that classifies, tags, or drafts a thousand times an hour. on-screen: A tool, not an oracle |
expected on screen: red ground · a JsonDiff panel over a dimmed Signal Field · Augur leads · code · crossfade-8f · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id JsonDiffvisual codeshape coneground redtreatment liquid-chromemotion crossfade-8fpower morphinstrument morph→A tool, not an oraclecurate codeLines, fileName, lines | |||||||
| 6 |  matches intent ChecklistCard template | teach | 37.0–44.3s | kinetic-build | icon·white-knockout | ♪ node_lock | Hand the model your own docs at call time so it answers from your facts, not its foggy memory. on-screen: Give it your notes to read |
expected on screen: red ground · a ChecklistCard panel over a dimmed Signal Field · Augur leads · node-graph · kinetic-build · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id ChecklistCardvisual node-graphshape coneground redtreatment white-knockoutmotion kinetic-buildpower morphinstrument morph+laser→Give it your notes to readcurate items, nodes | |||||||
| 7 |  matches intent StatScoreboard template | proof | 44.3–51.599999999999994s | receipts-count | icon·liquid-chrome | ♪ node_lock | Point it at a backlog of untagged tickets and it sorts every one before the office lights come on. on-screen: One night, a thousand tags |
expected on screen: red ground · a StatScoreboard panel over a dimmed Signal Field · Augur leads · receipts · receipts-count · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id StatScoreboardvisual receiptsshape coneground redtreatment liquid-chromemotion receipts-countpower receiptsinstrument spotlight→One night, a thousand tagscurate pillar, stats, statsLabels | |||||||
| 8 |  matches intent ProcessFlow template | step | 51.6–58.9s | kinetic-build | icon·white-knockout | ♪ node_lock | So the move is call from code, demand JSON, ground it in your docs, then run the batch overnight. on-screen: Step: call, JSON, ground, batch |
expected on screen: red ground · a ProcessFlow panel over a dimmed Signal Field · Augur leads · pipeline · kinetic-build · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id ProcessFlowvisual pipelineshape coneground redtreatment white-knockoutmotion kinetic-buildpower throwinstrument laser-trace→Step: call, JSON, ground, batchcurate stages, steps | |||||||
| 9 |  matches intent shared field signature-3d | resolve | 58.9–66.2s | coalescence | icon·liquid-chrome | ♪ bed_out | Treat AI as plumbing, not a party trick. The Big T-M wires it into the work you already do. on-screen: AI as plumbing, not party trick |
expected on screen: red ground · cone hero in the shared Signal Field · Augur leads · coalescence · coalescence · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape coneground redtreatment liquid-chromemotion coalescencepower coalescenceinstrument coalescence→AI as plumbing, not party trick | |||||||
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)
tiktokUsing AI for real, not as a chat toy (AI-assisted). Stop pasting prompts into a chat box. Call the model from your code, force the answer into strict JSON, and it plugs straight into your database. Hand it your own docs so it answers from your facts. Then point it at a backlog and it tags everything overnight.
instagramStop chatting. Start calling.
Call the model from code.
Force strict JSON output.
Ground it in your own docs.
Run the batch overnight.
#artificialintelligence #aiautomation #api #machinelearning #aitools
linkedinA few years ago, using AI meant a research lab and a patient budget. Now the same capability is one API call. The hard part moved from building models to using them well.
The shift that matters for teams:
1. Stop pasting prompts into a chat box. Call the model from your code so the work repeats reliably.
2. Demand strict JSON output so the response plugs straight into a spreadsheet or database, no copy-paste.
3. Ground it. Hand the model your own documents at call time so it answers from your facts, not its foggy memory.
The value isn't a smarter chatbot, it's a function that classifies, tags, or drafts a thousand times an hour. Point it at a backlog overnight and it's done by morning. Treat AI as plumbing, not a party trick. AI-assisted and disclosed.
xStop pasting prompts into a chat box. Call the model from code, force strict JSON, ground it in your own docs. Now it's a function that tags 1,000 tickets overnight, not a party trick. temerarii.com
facebookUsing AI for real doesn't mean chatting with a bot. Call the model from your code, force the answer into strict JSON so it plugs into your database, and hand it your own docs so it answers from your facts. Then point it at a backlog and it tags everything overnight. Treat AI as plumbing, not a party trick. Learn more at temerarii.com.
threadsStop pasting prompts into a chat box. Call the model from your code, force strict JSON, and ground it in your own docs. Now it's a function that tags 1,000 tickets overnight, not a party trick.
pinterestHow to actually use AI in your business: call the model from code, force strict JSON output, ground it in your own documents, and batch-process backlogs overnight. An evergreen guide to treating AI as reliable plumbing, not a chat toy.
blueskyStop pasting prompts into a chat box. Call the model from code, force strict JSON, ground it in your own docs. Now it tags 1,000 tickets overnight, not a party trick. temerarii.com
youtubeUsing AI For Real: API Calls, Strict JSON & Grounding in Your Own Docs
A few years ago, using AI meant a research lab and a patient budget. Now it's one API call, and the hard part is using it well. This covers the shift that matters: stop pasting prompts into a chat box and call the model from your code, demand strict JSON so the output plugs straight into a database, and ground it by handing the model your own documents so it answers from your facts. The value is a function that classifies and tags a thousand times an hour, not a smarter chatbot. AI-assisted and disclosed. More at temerarii.com.