← Back
AWS ML Blog

Natera’s intelligent appointment scheduling with Amazon Bedrock AgentCore

•20 min read•
#agents#bedrock#deployment#llm
Level:Intermediate
For:ML Engineers, RAG Practitioners
✦TL;DR

Natera engineered an automated voice agent on Amazon Bedrock AgentCore that enables patients to book mobile phlebotomy appointments via natural conversation. The system uses a dual‑WebSocket bridge to maintain real‑time interaction, employs event‑driven latency masking to hide backend delays, and implements progressive‑trust authentication to secure user data—all while achieving 100 % tool‑calling accuracy. This architecture demonstrates how Bedrock AgentCore can be leveraged for low‑latency, secure voice workflows in healthcare.

⚡ Key Takeaways

  • 100 % tool‑calling accuracy achieved in the appointment booking flow.
  • Dual‑WebSocket bridge architecture for bi‑directional, real‑time communication.
  • Event‑driven latency masking to conceal backend processing delays.
  • Progressive‑trust authentication integrated into the voice agent pipeline.
  • Deployment via Amazon Bedrock AgentCore with WebSocket endpoints.
  • WhyItMatters: Engineers building production voice assistants can adopt Bedrock AgentCore’s dual‑WebSocket and latency masking patterns to deliver responsive, secure, and fully automated user interactions.
  • TechnicalLevel: Intermediate
  • TargetAudience: ML Engineers, RAG Practitioners
  • PracticalSteps:
  • Configure a dual‑WebSocket bridge between the voice client and AgentCore.
  • Enable event‑driven latency masking in the AgentCore settings.
  • Implement progressive‑trust authentication before tool‑calling.
  • ToolsMentioned: Amazon Bedrock, AgentCore, WebSocket
  • Tags: AGENTS, BEDROCK, DEPLOYMENT, LLM

🔧 Tools & Libraries

Amazon BedrockAgentCoreWebSocket
💡 Why It Matters

Engineers building production voice assistants can adopt Bedrock AgentCore’s dual‑WebSocket and latency masking patterns to deliver responsive, secure, and fully automated user interactions.

✅ Practical Steps

  1. Configure a dual‑WebSocket bridge between the voice client and AgentCore.
  2. Enable event‑driven latency masking in the AgentCore settings.
  3. Implement progressive‑trust authentication before tool‑calling.

Want the full story? Read the original article.

Read on AWS ML Blog ↗

More like this

Censored vs. Truncated Regression in Python, Explained

Towards Data Science•#llm

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

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