---
title: "Roku AI Powered Support Chat"
description: "Technical lead for the front end and LLM routing on Roku's AI support chat, integrating Claude V2 and GPT-3/4 over the vendor's platform via AWS Connect. The vendor's published case study reports 48% AI containment, with chat rising from 30% to 43% of support contacts."
author: "Filipe Brito Ferreira (Senior Front-End Platform Engineer)"
canonical: https://www.fbritoferreira.com/projects/roku-ai-chatbot/
published: 2023
tags: ["AWS", "AI", "LLM", "RAG", "React", "AWS Connect", "Node.js"]
image: https://cdn.fbritoferreira.com/assets/projects/roku-chatbot-1-1200.webp
---

## What we built

I was the technical lead for the front end and the LLM routing/integration on an AI chatbot for Roku's customer support, built on the vendor's conversational platform. The system plugs into multiple LLM providers (Claude V2, GPT-3/4) through AWS Connect to handle customer queries with natural language understanding.

## Results

Per the vendor's published case study of the deployment:

- **48% AI containment**, resolved without a human agent
- Chat share of support contacts rose from **30% to 43%**
- Phone share fell from **70% to 57%**
- Response time went from hours (human queue) to seconds

## How it works

The platform pairs AWS Connect for comms infrastructure with a custom AI orchestration layer in React and Node.js:

- **Multi-LLM routing** picks the right model per query (Claude for nuanced reasoning, GPT for speed)
- **RAG pipeline** pulls from Roku's knowledge base so answers stay accurate and contextual
- **Escalation to humans** happens automatically for complex issues, with full conversation context passed along
- **Feedback loop** uses interaction data to keep improving resolution accuracy

## Technical details

- React chat widget on the frontend, Node.js orchestration layer, AWS Connect for the communication backbone
- Conversation context management keeps multi-turn dialogues coherent
- Fallback strategies make sure no query goes unanswered
- Monitoring dashboards track resolution rates, escalation patterns, and per-model performance

## In the press

Roku's AI transformation using the platform we built was covered by Quiq: [Roku's Agentic AI CX Transformation](https://quiq.com/blog/roku-agentic-ai-cx-transformation/).