Ask: Help Designing n8n Workflows for an AI Pet Adoption Agent (Data Ingestion + Screening + Notifications)
I’m building an AI assistant to help animal shelters screen adoption applications and match adopters with pets. I’d like guidance / example workflows in n8n for three things:
- Automated data collection from shelter/rescue websites
- Periodically scrape or ingest pet listings (name, age, size, breed, status, etc.) from different shelter websites.
- Handle that each site may have a different layout/CMS.
- Normalize the data into a consistent structure and store it in a database (e.g., Postgres or any DB n8n supports well).
- Customizable screening criteria + top 3 candidates per pet
- Let each shelter/location define its own screening rules (e.g., must-have questions, red flags, weighting for compatibility score).
- Use those criteria to score all applicants for a specific pet and automatically pick the top 3 candidates.
- Return this ranking in a clean JSON response so it can be consumed by a separate conversational front-end.
- Adopter preference storage + auto-notifications
- Store each adopter’s preferences (e.g., dog: age range, size, breed, energy level).
- When a new pet is ingested that matches those preferences, automatically:
- Log the match in the DB, and
- Trigger a notification hook (e.g., webhook, email, or message queue) so another system can notify the adopter.
I’m looking for best-practice patterns, node recommendations, and example n8n JSON workflows for:
- Web scraping / crawling varied websites
- Rule-based or score-based ranking flows
- Scheduling / polling and event-based triggers
- Designing the data model in a way that works well with n8n
I don’t need help with the conversational UI itself, just the n8n side of the automation.