Luis Rendon
AI Automation Engineer — n8n · WhatsApp AI · RAG Pipelines ·
New to Contra
Luis is building their profile!
Production WhatsApp AI agent built for a real client (PixelPages). The bot handles customer inquiries 24/7 using a RAG pipeline — it searches a vector knowledge base, generates context-aware responses via OpenAI, and automatically escalates to a human agent when it hits the boundaries of its knowledge. Self-hosted on Oracle Cloud with Docker, orchestrated with n8n, and running in production across multiple tenants with isolated data per client.
1
23
Production-grade n8n workflow that detects when an AI agent reaches the boundaries of its knowledge and automatically escalates to a human agent. The AI responds with a trigger phrase (ESCALATE_TO_HUMAN) when it cannot handle a request — an IF node detects it using contains logic with case-insensitive comparison, sends a holding message to the user, and simultaneously notifies the human agent with the session ID and original message. FALSE branch delivers the normal AI response untouched. Built from a real production WhatsApp AI system handling live customer conversations.
1
12
n8n workflow template that transcribes voice messages using OpenAI Whisper and automatically classifies the intent of the transcribed text using GPT-4o-mini. Routes the message to different branches based on detected intent — GREETING, QUESTION, REQUEST, or OTHER. Uses contains logic instead of equals for robust detection, and .trim().toUpperCase() to handle model output inconsistencies. Ready to connect to any webhook source — Evolution API, Twilio, or WhatsApp Cloud API. Published on the official n8n template marketplace.
1
10
n8n workflow that validates incoming WhatsApp phone numbers against a blacklist before passing messages to any AI or automation pipeline. Uses a JavaScript Code node with a hardcoded array for simplicity — replaceable with dynamic sources like Google Sheets, Airtable, or PostgreSQL for production deployments. Includes an IF node with boolean detection and two clear branches: blocked numbers receive an immediate rejection response, cleared numbers proceed to AI processing. Built from a real production multi-tenant WhatsApp AI system. Part of a series of production-ready WhatsApp automation building blocks.
1
17