Full-Stack Automation Engineer (Python, Web Scraping, Multi-Source Content, AI Trailer Tagging, BunnyCDN, WordPress API)

## Project Overview

I run a VR adult content website and currently manage a highly manual workflow involving downloading media, organizing files, generating content, uploading to a CDN, and preparing data for WordPress.

I am looking for an experienced developer to build a **fully automated pipeline** that replaces this manual process end-to-end while maintaining **trailer-specific, accurate tagging** and consistent SEO structure.

---

## Sources of Content

The system must handle content from multiple sources:

* **Primary aggregator website** – scrape pages and extract trailers, previews, and images
* **Premium AR sites** – direct download of trailers, previews, and images (no login required)
* **Affiliate panels / partner content** – provided via direct download links or CSVs (no login required)
* **Direct submissions** – zip files sent via email or stored locally

Once content is obtained, the processing pipeline (AI tagging, CDN upload, WordPress integration) should be **identical for all sources**.

---

## Input Control & Source Tracking

* Users will provide a **list of URLs** (e.g., CSV or text file) for the system to process
* Only provided URLs should be processed — no uncontrolled crawling
* For each video, the system must store:

* Original source URL
* Original title
* Studio name
* Extracted metadata (performers, etc.)
* Generated title and slug
* Tags and CDN URLs
* Prevent duplicate processing of the same URL
* Output stored in structured format (CSV and/or database) for easy tracking and future reference

---

## Scope of Work

### 1. Web Scraper & Downloader

* Scrape specified URLs from aggregator sites
* Download trailers, preview videos, and images from all sources
* Organize files into structured folders
* Apply consistent file naming

---

### 2. File Renaming & Image Processing

* Automatically generate SEO-friendly slugs
* Rename all media files accordingly
* Resize and optimize images

---

### 3. AI Content Generation & Trailer-Specific Tagging

* Generate:

* SEO-optimized titles
* HTML-formatted descriptions
* Tags from a predefined master list
* **Trailer-specific, hybrid approach**:

* Start with metadata as candidate tags
* Validate tags using frame extraction from trailers (~1 frame every 5–10 seconds)
* Flexible tag count: ~20–35+, fewer for simpler content (e.g., hentai/CGI)
* Ensure tags accurately reflect trailer content

---

### 4. BunnyCDN Upload Automation

* Upload media programmatically using BunnyCDN Storage API
* Generate correct public URLs for all media
* Support bulk/parallel uploads

---

### 5. Data Structuring / CSV Generator

* Automatically generate CSVs containing:

* Original URL and title
* Studio
* Your new title
* Slug
* Description
* Tags
* CDN URLs
* System should also optionally push data directly to WordPress via API

---

### 6. WordPress API Integration

* Upload media and posters
* Set alt tags automatically (format: title + “ - passthrough AR VR porn video - MyWebsite.com”)
* Create/schedule posts
* Associate metadata, images, and tags correctly

---

## Technical Requirements

* Python
* Web scraping frameworks (Playwright, Selenium, or similar)
* Media downloading (yt-dlp or similar)
* API integration: BunnyCDN + WordPress REST API
* Image processing (Pillow or similar)
* AI/ML integration (OpenAI, local CLIP models, or similar)
* Automation pipelines for multi-source input

---

## Deliverables

* Fully working automation system
* Clear documentation for setup and operation
* Scalable to ~100 videos per month
* Structured database or CSV output for tracking

---

## Screening Questions

1. Describe your experience with BunnyCDN Storage API and WordPress REST API
2. How would you implement trailer-specific AI tagging from a predefined list?
3. How would you handle multiple content sources and input-driven URL processing?
4. What tech stack and architecture would you use?

---

## To Apply

* Examples of relevant automation or scraping projects
* Technologies you would use
* Estimated timeline
* Suggestions for improvements to workflow