
Evaluating AI Agents: A production blueprint with Strands and AgentCore
Quick Answer
Motorway partnered with AWS to develop a dealer stock search agent using Strands Agents SDK and Amazon Bedrock AgentCore, reducing incorrect query results from 1 in 8 to 1 in 50 and cutting issue detection time from hours to minutes.
Quick Take
The solution leverages a robust evaluation pipeline and costs approximately $5–10 for running the sample evaluation suite.
Key Points
- The evaluation pipeline combines Strands Agents SDK with Amazon Bedrock AgentCore.
- A two-phase strategy improves agent reliability and reduces incorrect results significantly.
- The agent supports natural language queries, enhancing dealer search efficiency.
- Initial deployment takes 30-45 minutes, with customization requiring 2-3 hours.
- Security measures include least-privilege IAM roles and API key management.
DeepSignal Analysis
What happened
Motorway collaborated with AWS to create an AI dealer stock search agent, significantly improving search accuracy and reducing issue detection time. The evaluation pipeline they developed decreased incorrect query results from 1 in 8 to 1 in 50, and the time to detect issues was reduced from hours to minutes. The solution utilizes the Strands Agents SDK and Amazon Bedrock AgentCore.
Key evidence
- Motorway's AI agent reduced incorrect search results from 1 in 8 queries to 1 in 50, enhancing dealer trust.
- The evaluation pipeline cut issue detection time from a few hours to just a few minutes, improving operational efficiency.
- The estimated cost for running the sample evaluation suite is approximately $5–10 in Amazon Bedrock inference charges.
Why it matters
The development of this AI agent addresses critical challenges in the automotive marketplace, where accurate and timely information is essential for dealer operations. By improving search accuracy and reducing response times, Motorway enhances user trust and operational efficiency. This case illustrates the potential of AI in transforming traditional workflows, particularly in high-stakes environments like vehicle auctions.
Source Excerpt
Together, Motorway and AWS built an end-to-end evaluation pipeline that reduced incorrect results from 1 in 8 queries to 1 in 50 and cut issue detection time from few hours to few minutes. The pipeline combines the Strands Agents SDK with Amazon Bedrock AgentCore, a fully managed service for deploying and operating AI agents at scale. In this post, you will learn how to build this pipeline for your own agents.
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from AWS Machine Learning
See more →
AI Teammates: how monday.com runs production AI agents on Amazon Bedrock
monday.com leverages Amazon Bedrock to run AI agents at scale, achieving over 50% increase in per-engineer PR throughput. Their architecture integrates multiple AWS services, enabling seamless collaboration between human engineers and AI teammates across a decade-old code base.

