The AI stack on AWS has never been stronger. Amazon Bedrock gives teams access to leading models and the Strands Agents SDK makes agents easier to build. Amazon Bedrock AgentCore provides a managed runtime for operating them.
Each layer does its job, and one production question still spans all of them: will the application actually finish what it started when the model calls time out, tools fail, infrastructure restarts, or a workflow waits for hours or days for a person to respond?
AI agents are long-running distributed systems. They call models and tools, coordinate with other agents, cross service boundaries, and carry business state through every step. Once those systems matter to customers or the business, durable execution becomes a foundational requirement.
That is the role Temporal was built to play, and customers are the ones pulling AWS and Temporal together to do it. The result is a partnership that is deepening on two fronts: how AWS takes agentic AI to market with Temporal, and how the two companies’ products fit together across the major layers of the agent stack.
AWS is building Temporal into its AI programs#
In November 2025, AWS launched its largest specialization expansion to date, introducing the Agentic AI categories alongside a new Resilience Software Competency for Technology Partners.
The Agentic AI Tools Competencies validate technical depth in agentic systems. The Resilience Software Competency validates the operational rigor to keep production workloads highly available.
Temporal achieved both: the Agentic AI Tools competency in April 2026, and Resilience Software Competency in September 2026. Together, they say something more specific than either one alone: Temporal brings the operational reliability that keeps agentic systems running in production.
AWS has made durable execution a core component of its education series on agent development. The AWS Marketplace Agentic AI Learning Series presents it as a foundation for reliable agent workflows and features Temporal Cloud as a code-native orchestration option. AWS is placing durability in the architecture conversation from the beginning, and explicitly identifying Temporal as part of that conversation.
The partnership’s momentum reflects a track record of successful joint customer engagements. Together, AWS and Temporal have helped customers modernize critical applications and realize meaningful value from AI, supported by close collaboration across both companies’ product and go-to-market teams.
That combination of customer outcomes, technical collaboration, and commercial success led AWS and Temporal to renew and expand their multi-year Strategic Collaboration Agreement two years ahead of schedule. The expanded agreement builds on that foundation, supporting deeper collaboration to bring joint solutions to more customers.
Deeper in the AWS AI stack#
AWS and Temporal have integrations across the major layers of an agentic application.
At the agent framework layer, Temporal’s TypeScript and Python integrations for the Strands Agents SDK routes model invocations, tool calls, MCP calls, and hooks through Temporal Activities. Developers keep the Strands programming model while Temporal makes the execution recoverable and observable, and lets it pause and resume safely.
At the compute layer, Serverless Workers on AWS Lambda and Amazon Bedrock AgentCore Runtime are now in Public Preview and Pre-release respectively. Temporal Cloud invokes Workers when there is demand and compute scales back to zero when it is idle. Teams get durable workflow state without maintaining a Worker fleet around the clock, a natural fit for event-driven and bursty AI workloads.
AWS and Temporal published a joint architecture and sample showing Temporal orchestrating agents hosted on Amazon Bedrock AgentCore. A global consumer packaged goods company has already used this pattern to reduce backend infrastructure costs by 66% while processing millions of events per month.
Across technical validation, developer education, joint customer engagements, and product integrations, the collaboration now connects how teams learn, build, and bring critical applications and AI into production on AWS.
Explore what AWS and Temporal can help you build#
Modernizing a critical application or taking an AI workflow into production? Bring your use case to Temporal or your AWS account team to explore how durable execution can help.
Start with the joint Temporal and Amazon Bedrock AgentCore architecture and sample, or connect with Temporal to discuss your workload.
Learn more in person#
You can meet the Temporal and AWS teams at these upcoming events.
- October 29: Join AWS, Temporal, Arize AI, and Databricks at the AWS Builder Loft in San Francisco for an Agentic AI Partner Showcase focused on taking agents into production. RSVP here.
- November 30 to December 4: Meet Temporal at AWS re:Invent in Las Vegas, where Temporal will be a Platinum sponsor with new product news, technical sessions, and demos. Book a meeting with Temporal experts here.
- Available now: Watch Why Production Agents Fail and How to Fix It with Temporal and AWS on demand, or explore Temporal Cloud in AWS Marketplace.