Browse categories
Explore
Fiverr Pro
English
$
USD
Looks Like This Service Is On Hold
I will build a whatsapp ai agent with n8n and rag
Lukman Indro P
About this gig
Your business is losing leads because no one responds to WhatsApp messages fast enough or at all after business hours.
I build WhatsApp AI agents using n8n, Gemini, and RAG pipelines backed by PGVector. The agent answers questions from your knowledge base, captures lead data, and triggers automated follow-up messages on a schedule you define. I have deployed this in production for a real client cutting response time from minutes to 10 seconds and increasing lead conversion from under 20% to over 35%.
At the end of the project, you get a working n8n workflow connected to your WhatsApp number, a configured knowledge base your team can update without touching code, and documented handoff so you are not dependent on me to keep it running. Higher tiers include multi-agent architecture with guardrail screening, quality checking, and an admin fallback loop for questions the AI cannot answer.
Message me before ordering if you're not sure which package fits your project.
Get to know Lukman Indro P
Lukman Indro P
AI Automation Engineer
- FromIndonesia
- Member sinceJan 2018
Languages
English, Indonesian
AI Automation Engineer and LLM Integration Specialist with 2+ years building production LLM-powered pipelines, agentic workflows, and API automation systems for business clients. Shipped products spanning n8n automation workflows, RAG pipelines, and multi-agent architectures — including a legal-tech contract audit system and a WhatsApp AI agent with measurable conversion impact. Co-founded an EdTech platform scaled to 50,000 users. Works independently in remote environments with clear system documentation for cross-functional handoff.
My Portfolio
FAQ
What WhatsApp API do you use, and do I need to pay for it separately?
I build with WAHA (WhatsApp HTTP API), which you self-host. It is not the official WhatsApp Business Cloud API. If your business already uses or requires the official Meta API, tell me before ordering — the integration approach differs and this depends on your project.
Can I update the chatbot's knowledge base myself after delivery?
Yes. For Starter and Standard tiers, the knowledge base is Google Sheets — your team can add or edit rows without touching the workflow. For Premium, new entries auto-sync into the vector store. I document this process in the handoff.
What happens if the AI cannot answer a question?
In the Standard tier, unanswered questions are logged. In the Premium tier, the system automatically notifies your admin via WhatsApp and delivers the admin's reply back to the original user. The Starter tier does not include this fallback.
