From failed build to AI-guided fix
A silent screen demo turns a failed deploy into an AI diagnosis, one-line correction, and verified success.
Field notes / why it works
The opening diagnostic button poses the question before 0:01, then the build visibly fails by about 0:08. The assistant's specific typo diagnosis at 0:14–0:17, code change at 0:20, and completed deploy by 0:26 turn a feature claim into a short, verifiable workflow; the opening crop returns at 0:27.6 as a visual loop.
Plate 01 / The format in one breath
A silent, horizontal product walkthrough begins on an oversized AI troubleshooting button, then shows a failed build, a plain-language diagnosis, the one-line code correction, and a successful deploy. The end returns to the same button graphic, forming a visual loop.
Why it works
- The first frame asks “Why did it fail?” over a blurred log; the problem and promised tool are visible before 0:01.
- It offers concrete proof: a failed build at 0:08–0:10, a specific typo and proposed correction at 0:14–0:17, a code edit at 0:20, and completed deploy at 0:24–0:27.
- Crops move from button to full workflow to close diagnosis, letting viewers read the important detail without a narrator; there are 12 measured picture changes across 31.6 seconds.
Plate 02 / Format card
| Platform · aspect · length | X · 16:9 · 1280×720 · 31.6 s · 30 fps |
| Pace | 3 shots, 2 definite cuts; 3.8–5.7 cuts/min including one uncertain change; 10 internal steps; 22.7 total picture changes/min; mean shot 10.55 s |
| Script | No dependable spoken script. The automated Korean transcript repeats a syllable at 743 wpm, inconsistent with a legible narration; do not use it as dialogue. |
| Voice | None verified; interface text tells the story. |
| On screen | Approximately 100% screen recording and UI crops: deploy log, AI diagnosis, code editor, success page. |
| Captions | No editorial subtitles; original UI labels and diagnosis are the reading layer. |
| Sound | Extremely quiet measured track (−45.7 LUFS integrated); music and effects cannot be identified reliably from the supplied evidence. |
| Structure | Button hook → failed status → AI diagnosis → exact fix → verified success → return to button; no in-video CTA. |
Pro / 13 plates + recipe
The first plates are open. The full dissection is Pro.
Get the timed shots, type and sound specs, and the runnable recipe.
Open the full blueprint- Plate 03 / The hook, frame by frame (first 3 seconds)Pro
- Plate 04 / Structure (beat sheet)Pro
- Plate 05 / Shot grammarPro
- Plate 06 / Visual style guidePro
- Plate 07 / Voice and deliveryPro
- Plate 08 / Sound designPro
- Plate 09 / PackagingPro
- Plate 10 / Script templatePro
- Plate 11 / Production planPro
- Plate 12 / PromptsPro
- Plate 13 / Edit recipePro
- Plate 14 / QA checklistPro
- Plate 15 / Make it yoursPro
- Recipe / runnable promptPro