AWS Shows an Agentic Approach to Conversational Video Intelligence
AWS outlined a video-question-answering solution in which one Strands Agents SDK agent coordinates Amazon Bedrock, Amazon Rekognition, and Amazon Transcribe at runtime, choosing which service to call as users ask natural-language questions about their videos.

Full Report
On September 23, 2026, AWS published an approach to building conversational video intelligence on its cloud platform. A single agent built with the Strands Agents SDK orchestrates Amazon Bedrock, Amazon Rekognition, and Amazon Transcribe at runtime, deciding which service to call for a given request.
The solution lets users ask natural-language questions about video and is described as returning answers in seconds. By making service selection part of the agent’s runtime behavior, the design brings the three services into one video-questioning workflow.
The post adds another example to AWS’s recent agentic-workflow demonstrations. On September 29, 2026, AWS separately described a synthetic-monitoring architecture using Amazon Nova Act and Amazon Bedrock AgentCore to run scheduled browser-based user journeys and alert teams when a journey fails. The examples address different tasks—video questions and browser monitoring—while both show agent-based workflows built around AWS services.
Why This Matters
For teams that need to find information in video, natural-language queries and runtime service selection offer a way to connect multiple AWS services in one interactive workflow. The example also gives readers a concrete comparison with AWS’s separate agent-based approach to scheduled browser monitoring.
Evidence
Start with the primary evidence, then review supporting sources and context.
Primary Evidence
1Agentic conversational video intelligence built on AWS
Michael Li
“Learn how to build a conversational video intelligence solution on AWS using an agentic architecture. A single Strands Agents SDK agent orchestrates Amazon Bedrock, Amazon Rekognition, and Amazon Transcribe at runtime, deciding which service to call so you can ask natural language questions about your videos and get answers in seconds.”
Recommended signals
Recommended by brand, topic, and recent related coverage