Wallpaper Product Catalog Data Pipeline, n8n Workflow, LLM Processing & Cloud Run Service
Budget: $200.0
FIXED /
⭐ 4.98 (21)
Israel
data-scraping, python
We're looking for a developer to build a complete product data pipeline for our wallpaper catalog. All 5 components are bundled into a single fixed-price contract of $200 USD, paid as one milestone upon completion of all of them.
What needs to be done:
1) Cloud Run Scraper Service Setup:
Set up a reliable Python-based scraper service deployed on Google Cloud Run. The service should be able to retrieve product pages from our own website catalog reliably and efficiently.
2) Python Scraper for Wallpaper Product Catalog:
Build a Python scraper that collects all wallpaper product data from our catalog. The scraper should handle pagination and extract all relevant product fields (name, images, dimensions, SKU, category, etc.).
3) n8n Automation Workflow:
Create an n8n workflow that orchestrates the data pipeline end to end: triggering the scraper, collecting output, and passing data to downstream steps (LLM processing, JSON generation, and storage).
4) LLM-Based Name Translation and Data Processing:
Integrate an LLM step (e.g. OpenAI or similar) to translate product names and normalize the data. Output should be clean, structured product records ready for import.
5) JSON Output Generation and BE Service:
Generate a standardized JSON file of all collected and processed products. The backend service on Cloud Run should expose this data and include logic to detect changes between runs (e.g. new products, updated fields, removed items).
Deliverables:
- All services deployed and running on Google Cloud Run.
- n8n workflow exported and documented.
- JSON output file with full product catalog.
- BE service with change-detection logic.
- A short handover note explaining the architecture, environment variables, and how to re-trigger the pipeline.
Timeline: roughly 7 to 10 business days total.
Revisions: up to 2 rounds of revisions per component within original scope, included.
IP and Confidentiality: All code becomes our property upon final payment. Contractor agrees not to share or reuse the codebase or business logic.
Ideal candidate: intermediate-to-senior Python developer comfortable with data pipelines, n8n, LLM APIs, and Google Cloud Run deployments.
When applying, please briefly mention similar data pipeline or automation work you have done.
Öppna på Upwork