← Back
AWS ML Blog

Building a restaurant telephony AI host with Amazon Connect

•21 min read•
#agents#amazon#deployment
Level:Intermediate
For:AI Engineers
✦TL;DR

This article demonstrates how to build a voice‑ordering system for restaurants using Amazon Connect Customer for telephony, Amazon Lex V2 with Amazon Connect Agentic Voice for real‑time speech, and an Amazon Connect AI agent orchestrated via the Model Context Protocol (MCP) and Amazon Bedrock AgentCore. The architecture cleanly separates the call handling, conversational reasoning, and backend order management, allowing the backend to evolve without modifying the agent. Engineers can deploy the entire stack with AWS CDK, enabling a fully managed telephony‑first AI ordering experience.

⚡ Key Takeaways

  • The solution uses Amazon Connect Customer to receive inbound phone calls and route them through a contact flow.
  • Conversation logic runs in an Amazon Connect AI agent while speech is handled by Amazon Connect Agentic Voice (Advanced ASR and TTS).
  • Backend services are exposed as discoverable tools via AgentCore Gateway and the open‑standard Model Context Protocol, so backend changes don’t require agent rewrites.
  • Deploy the full stack with AWS Cloud Development Kit (AWS CDK) to provision Amazon Connect, Lex V2, and the AI agent in a single deployment.
  • The implementation is limited to the telephony channel; it does not support web or app‑based ordering flows.
💡 Why It Matters

By leveraging fully managed AWS services and MCP, engineers can add AI‑driven phone ordering to existing restaurant operations with minimal custom infrastructure, reducing manual order taking and wait times.

✅ Practical Steps

  1. Use AWS CDK to deploy Amazon Connect Customer, the contact flow, Amazon Lex V2 bot, and the Amazon Connect AI agent as described in the walkthrough.
  2. Configure the contact flow to open an Amazon Connect AI agent session and route the caller through Amazon Connect Agentic Voice for speech recognition and synthesis.
  3. Connect the AI agent to your restaurant backend by exposing backend APIs as MCP tools through the AgentCore Gateway.

Want the full story? Read the original article.

Read on AWS ML Blog ↗

More like this

New in AWS Quick: More capable flows, an updated web experience, and agents that work together

AWS ML Blog•#agents

Three insights you may have missed from theCUBE’s coverage of the Dell Technologies New York Symposium

SiliconANGLE AI•#deployment

How Google Duplex evolved to also call businesses for you in Gemini

Google Research•#agents

Welcome Inkling 2 Mini, Cutting-edge Performance for your Local Setup

Hugging Face Blog•#llm

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING