Follow
Follow
AI AUTOMATION

Multimodal GenAI Support Bot

Year
2026

Overview

An e-commerce and SaaS brand needed an intelligent, 24/7 autonomous support agent to capture off-hours leads, resolve complex customer tickets, and track user sentiment.

Impact

The client was losing high-intent leads and frustrating customers because their human support team was offline after hours. Their existing rule-based chatbot was rigid, unable to answer nuanced questions from their documentation, and lacked the ability to gauge customer frustration, leading to escalated churn and lost revenue.

Work

I developed and deployed a multimodal AI Customer Service Bot using n8n and OpenAI. Moving far beyond basic decision trees, this “Delegate Agent” utilizes Retrieval-Augmented Generation (RAG) to instantly answer complex queries based on the company’s proprietary internal documents. Furthermore, I engineered it to run real-time sentiment analysis, automatically escalating the chat and pinging the human team on Slack the moment a customer became highly frustrated.

70% Tier 1 Resolution

Autonomously resolved the vast majority of standard support tickets without human intervention, drastically clearing the manual support queue.

Real-Time Sentiment Routing

Successfully engineered the bot to identify and escalate high-risk, frustrated customers to human agents instantly via Slack, preventing churn.

24/7 Lead Capture

Acted as a continuous, intelligent touchpoint to qualify and capture high-intent leads outside of standard business operating hours.