---
title: "Discussion on \"An AI video prompt pre-flight checklist: 12 checks before you spend a generation\" | Hashnode"
description: "Discussion on \"An AI video prompt pre-flight checklist: 12 checks before you spend a generation\". Most wasted generations come from four decisions made before the model sees the prompt."
url: "https://hashnode.com/posts/ai-video-prompt-checklist/6ae2daa0d1cd3848f7437aa6"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [An AI video prompt pre-flight checklist: 12 checks before you spend a generation](https://callirra.hashnode.dev/ai-video-prompt-checklist?utm_source=hashnode&utm_medium=feed)

Most wasted generations come from four decisions made before the model sees the prompt.

Oct 29 · 6 min read · callirra.hashnode.dev  
Read the article

-   [#prompt-engineering](https://hashnode.com/tag/prompt-engineering)
-   [#text-to-video](https://hashnode.com/tag/text-to-video)
-   [#ai-video](https://hashnode.com/tag/ai-video)
-   [#video-generation](https://hashnode.com/tag/video-generation)
-   [#image-to-video](https://hashnode.com/tag/image-to-video)

## Written by

[Callirra](https://hashnode.com/@callirra)

@callirra

  
[View profile](https://hashnode.com/@callirra)

One workspace for AI image, video and audio

Callirra is one workspace for AI image, video and audio: 50 media models, 11 multimodal chat models, prompt polish, a private asset library, saved Constellation workflows and AI Filmmaker. The team writes here about real credit costs, model comparisons and release notes. Product and pricing: callirra.com

## Responses 4

Markdown supported · ⌘↵ to post

[Liam O'Sullivan](https://hashnode.com/@liamosullivan)

[Oct 29](https://hashnode.com/posts/ai-video-prompt-checklist/6ae2daa0d1cd3848f7437aa6/comment/6ae2e2cb3e9c8a6d5115cb83)

The reference-ratio problem shows up in motion work too. We keep a batch of pre-cropped references per output format, which cut retries noticeably.

[Priya Raman](https://hashnode.com/@priyaraman)

[Oct 29](https://hashnode.com/posts/ai-video-prompt-checklist/6ae2daa0d1cd3848f7437aa6/comment/6ae2e23027cf9dfd81c93141)

Curious how you handle the aspect-ratio mismatch when the reference is a phone clip. Do you crop first, or let the model re-frame and accept the drift?

[Callirra](https://hashnode.com/@callirra)

[Oct 29](https://hashnode.com/posts/ai-video-prompt-checklist/6ae2daa0d1cd3848f7437aa6/comment/6ae2e23027cf9dfd81c93141#reply-6ae2e4573e9c8a6d5115d96c)

We crop first when the subject already fills roughly 60% of the frame — a straight crop keeps framing predictable. Below that we let the model re-frame, because cropping a wide shot into 9:16 tends to slice limbs or on-screen text at the edges. The drift is real though, so we constrain subject position in the prompt and re-roll rather than try to fix it in post.

[Sofia Marchetti](https://hashnode.com/@sofiamarchetti)

[Oct 29](https://hashnode.com/posts/ai-video-prompt-checklist/6ae2daa0d1cd3848f7437aa6/comment/6ae2ecf1662d5ddbb1829892)

Bookmarking this before our next batch run.
