Postman details how Agent Mode runs on Amazon Bedrock for 40 million developers
Overview
Postman has published an account of the architecture behind Agent Mode, its AI agent for API testing, documentation, discovery, and implementation, as described in an AWS Machine Learning Blog post.
According to the post, Postman limits the tools visible to the agent for each task, uses schema-based queries, and builds purpose-shaped context handlers, while running on Amazon Bedrock with cross-Region inference and prompt caching. Postman reports that tool-selection errors rose once the visible toolset exceeded about 40 tools, which is the basis for its per-task tool limits. The source is a first-party account, so the design choices and the 40-tool threshold are Postman's own reported findings rather than independently verified results.
Written by AI from the articles below · updated Oct 9, 11:52 AM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
- AWS Machine Learning BlogPickHow Postman runs Agent Mode for 40 million developers on Amazon Bedrock
AIPostman describes the architecture behind Agent Mode, its AI agent for API testing, documentation, discovery, and implementation. The post covers limiting tools per task, using schema-based queries, building purpose-shaped context handlers, and running on Amazon Bedrock with cross-Region inference and prompt caching. Postman reports that tool-selection errors rose once the visible toolset exceeded about 40 tools.
Heat trend
Not enough continuous observations to show a trend yet.