Checklist the per-video bar — engine/sim
84.5/100
quantitative quality · weights learn from your reviews (
engine.sim.memory review thread-build-log-W49-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 · 36.3s · 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 | 0–8.2s | spatial-parallax | icon·white-knockout | ♪ bed_in | Week twenty-seven. We shipped pipeline simulations this week, four dry-runs that catch problems before we spend a cent on APIs. on-screen: Shipped: pipeline dry-runs |
expected on screen: red ground · knot hero in the shared Signal Field · Magister leads · node-graph · spatial-parallax · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual node-graphshape knotground redtreatment white-knockoutmotion spatial-parallaxpower summoninstrument summon→Shipped: pipeline dry-runs | |||||||
| 2 |  matches intent DiffCard template | build | 8.2–13.299999999999999s | type-on | icon·white-knockout | ♪ node_lock | The old way was discovering a broken prompt after you'd already paid for the generation. on-screen: Old way: find out after you pay |
expected on screen: red ground · a DiffCard panel over a dimmed Signal Field · Magister leads · code · type-on · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id DiffCardvisual codeshape knotground redtreatment white-knockoutmotion type-onpower summoninstrument draw-on→Old way: find out after you paycurate codeLines, lines | |||||||
| 3 |  matches intent TerminalRun template | diff | 13.3–20.200000000000003s | crossfade-8f | icon·liquid-chrome | ♪ node_lock | It broke on realism. The simulation passed but the real run failed, because the fake data was too clean. on-screen: What broke: the sim diverged from real |
expected on screen: red ground · a TerminalRun panel over a dimmed Signal Field · Magister leads · code · crossfade-8f · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id TerminalRunvisual codeshape knotground redtreatment liquid-chromemotion crossfade-8fpower morphinstrument morph→What broke: the sim diverged from realcurate codeLines | |||||||
| 4 |  matches intent AnnotatedDiagram template | step | 20.2–26.299999999999997s | kinetic-build | icon·white-knockout | ♪ node_lock | Now we catch failures for free. Next, we block any paid run until the simulation passes. on-screen: Next: gate spend on a clean sim |
expected on screen: red ground · a AnnotatedDiagram panel over a dimmed Signal Field · Magister leads · pipeline · kinetic-build · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id AnnotatedDiagramvisual pipelineshape knotground redtreatment white-knockoutmotion kinetic-buildpower throwinstrument laser-trace→Next: gate spend on a clean simcurate callouts, stages, steps | |||||||
| 5 |  first render · fix pending KpiGrid templatedead_airgeneric_scene | proof | 26.3–31.3s | receipts-count | icon·liquid-chrome | ♪ node_lock | Build Log |
expected on screen: red ground · a KpiGrid panel over a dimmed Signal Field · Magister leads · receipts · receipts-count · icon·liquid-chrome logo · caption bottom-left spec (the prompt): comp_id KpiGridvisual receiptsshape knotground redtreatment liquid-chromemotion receipts-countpower receiptsinstrument spotlight→Build Logcurate kpis, statsLabels | |||||||
| 6 |  first render · fix pending shared field signature-3ddead_airgeneric_scene | resolve | 31.3–36.3s | coalescence | icon·white-knockout | ♪ bed_out | Build Log |
expected on screen: red ground · knot hero in the shared Signal Field · Magister leads · coalescence · coalescence · icon·white-knockout logo · caption bottom-left spec (the prompt): comp_id shared Signal Fieldvisual coalescenceshape knotground redtreatment white-knockoutmotion 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: we shipped pipeline dry-runs this week. Four test runs catch problems before we spend a cent on APIs.
Old way: find out a prompt was broken after you already paid.
Next: block any paid run until the test run passes. (AI-assisted)
instagramBuild log this week.
Shipped: pipeline dry-runs.
Four test runs catch breaks before we pay for APIs.
It broke on realism: fake data was too clean.
Next: block paid runs until the test passes.
#buildinpublic #pipeline #automation #devlog #testing
linkedinBuild log, this week.
We shipped pipeline simulations: four dry-runs that catch problems before we spend a cent on APIs.
The old way was discovering a broken prompt after you had already paid for the generation.
What broke: realism. The simulation passed but the real run failed, because the fake data was too clean.
Next: we block any paid run until the simulation passes. We catch the failures for free now.
xBuild log: shipped pipeline dry-runs. Four test runs catch breaks before we pay for APIs. It broke when fake data was too clean. Next: block any paid run until the sim passes. temerarii.xyz
facebookBuild log for the week: we shipped pipeline dry-runs, four test runs that catch problems before we spend a cent on APIs. The old way was finding a broken prompt after you already paid. It broke on realism once because the fake data was too clean. Next up, we block any paid run until the test run passes. Follow along at temerarii.xyz.
threadsBuild log this week.
Shipped: four pipeline dry-runs that catch breaks before we pay for any API call.
Old way: find the broken prompt after you already paid.
It once broke because the fake data was too clean. Next: block any paid run until the sim passes.
pinterestBuild log: shipping pipeline dry-runs to cut wasted API spend. Four test runs catch broken prompts before any paid generation. A lesson learned: simulations can pass while the real run fails when the fake test data is too clean. The fix is to block any paid run until the simulation passes against realistic data.
blueskyBuild log: shipped four pipeline dry-runs that catch breaks before we pay for APIs. It once broke because fake data was too clean. Next: block any paid run until the sim passes.
youtubeTitle: Build Log: Pipeline Dry-Runs That Catch Breaks Before You Pay
This week we shipped pipeline simulations, four dry-runs that catch problems before we spend a cent on APIs. The old way was discovering a broken prompt after you already paid for the generation. It broke once on realism: the simulation passed but the real run failed because the fake data was too clean. Next, we block any paid run until the simulation passes. More at temerarii.xyz.