Turn a source video into several short social clips by specifying the audience, moments, pacing, and export format.
Source episode · 32:03
How I AI: My GPT-5.6 Sol Benchmark & 4 Game-Changing Workflows (vs. Fable)with Claire Vo
32:03 to 33:46: Claire drops a video into the model, asks for five social clips with sharp funny moments, and refines the edit for pacing and tone.
A set of distinct social-ready clips with accurate cuts, readable captions, and clean exports.
OpenAI's cloud-based AI software engineering agent that can execute code, run tests, and handle complex multi-file tasks autonomously.
Step by step
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
5 steps
Open a multimodal AI tool like Codex that accepts video files. Drag and drop your long form video, such as a talk or webinar recording, into the interface.
Upload the source video and transcript without recompressing the original.
Provide a simple, direct instruction in the chat interface. Specify the number of clips you want and their intended use, such as for social media, to guide the initial edit.
Create five clips for [platform], each [duration], aimed at [audience], using sharp, funny, high-energy moments with a complete payoff.
Review the clips the AI generates and provide follow up feedback in the chat to refine them. For example, ask for changes to the orientation, pacing, or editing, such as making the clips "horizontal, faster, and tighter."
Tighten this clip around [moment], keep the necessary setup, remove pauses, and preserve the speaker’s exact meaning.
Once the AI generates clips that meet your needs, download the video files from the tool.
Export approved clips in [aspect ratio, resolution, codec] with normalized audio and burned or sidecar captions.
For a more polished look, import the clips into a video editor like CapCut to add music, captions, or other branding elements before publishing.
Add the approved title card, caption styling, safe margins, and final audio check.
Turn an idea into a PRD, user stories, and a plan.
After the steps
How to recover when the loop fails and where human judgment helps.
Recover
Keep building
Use a hybrid AI workflow to bridge design and development. Generate high-level design concepts and tokens with Claude, then use Codex for robust implementation of the technical components in your codebase.
Implement a QA loop for AI-generated code using Devin. This playbook combines automated code reviews with an agent-recorded video walkthrough for visual verification, ensuring high-quality merges at scale.
Build a reusable 'Watchdog' playbook in Devin to automatically monitor customer accounts. The agent proactively checks for errors, synthesizes a prioritized list of critical issues, and saves you time.
Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.