Giving an AI agent capabilities has become remarkably easy. Giving one responsibility is much harder.
When agents begin talking to customers, operating production systems, or changing business data, a capable agent loop is only part of the solution. The surrounding application must also enforce policies, survive failures, preserve context across long-running interactions, support human approval, and provide a trustworthy record of what happened.
The Temporal Agent Harness is an outer harness designed for those requirements. It works with agent SDKs such as the OpenAI Agents SDK, PydanticAI, and Gemini, adding durable execution and application-level controls without replacing the tools and models developers already use.
In this webinar, we’ll introduce the ideas behind the Temporal Agent Harness and show how it treats agents as long-running application components rather than short-lived chat loops. We’ll explore durable turns, controls around tool execution, typed agent interfaces, human-in-the-loop interactions, and the event stream that records an agent’s trajectory as it works.
Can't make the live session? Register anyway, and we'll send the recording to your inbox.
Principal Developer Advocate
Temporal Technologies
Cornelia has spent a career at the forefront of technological innovation, starting with image processing algorithm development, moving to web-centric computing in the late 1990s, and then more than a decade working in cloud-native software and DevOps platforms. As Senior Staff Developer Advocate for Temporal, she is now helping to drive the expansion of the “durable execution” distributed systems paradigm.
She is the author of Cloud Native Patterns: Designing change tolerant software.
