Urgently Hiring - AI-Powered Mood Board Automation Tool for a Video Editor (Python + GPT-4 Vision + n8n)

looking for an experienced Python developer to build an end-to-end automation pipeline that streamlines how my client captures the visual inspiration from YouTube videos.

Currently, the workflow involves manually scrubbing through YouTube videos, pausing on frames that catches eye, and taking screenshots to save to mood board. This is time-consuming and entirely manual. this process needs to be automated.

The system should accept a YouTube video URL or channel name as input, intelligently extract the most visually compelling frames from the video using AI, and save those frames as high-quality images into an organised mood board folder.

What the System Needs to Do:-

The pipeline has four main stages working together:

Stage 1 — Input & Trigger. The system should accept either a direct YouTube video link or a YouTube channel name. I will trigger it through a simple n8n webhook, so I can run the whole thing just by submitting a URL.

Stage 2 — Video Download & Scene Extraction. Using yt-dlp, the tool downloads the video (video only, no audio needed). It then uses PySceneDetect to find every point where the visual scene changes, and extracts one representative frame per scene. This is the efficiency trick that avoids processing thousands of near-identical frames.

Stage 3 — AI Frame Selection (the core feature). Surviving candidate frames are filtered for blur and bad exposure using OpenCV, then stitched into a numbered contact sheet grid. That grid image is sent to GPT-4o Vision with a detailed style prompt describing the visual preferences (cinematic composition, dramatic lighting, strong colour grading, etc.). GPT reviews the grid and returns a list of the best frame numbers. This is what makes the tool intelligent rather than just mechanical.

Stage 4 — Output & Storage. The selected frames are extracted from the original video at full resolution and saved as high-quality JPEGs. These are then automatically uploaded to a designated Google Drive folder via n8n's Google Drive integration.

Project Scope & Deliverables:-

This is a fixed-scope project. Here is exactly what I expect at handoff:

A production-ready Python script (refined from the base I provide) with full error handling and logging.
A working n8n workflow, imported, connected, and tested — ready to receive a webhook and deliver files to Google Drive.
A short setup and maintenance guide (plain English, no jargon).
A test report showing the pipeline working correctly on at least 5 videos.

Budget & Timeline:-

I am looking to complete this project within 4 days. Please include your estimated timeline in your proposal.

For budget, I am open to hearing your rate. Fixed-price proposals are preferred. When applying, please give me a rough breakdown of where you expect to spend most of your time (e.g. deployment, n8n integration, testing) so I can understand your approach.

7 comments