[Workshop]
Building Durable, Production-ready Agents with Spring AI and Temporal
Cornelia Davis - Temporal / Marcus Merrell - Temporal
Everyone is building AI Agents, and looking for ways to build them more easily. Spring AI has already established itself as the premier Java framework providing an idiomatic way for Java developers to integrate LLMs, tools, and retrieval into their applications. More recently, Embabel has emerged to codify higher-level patterns for building goal-driven AI agents on top of Spring AI. With Spring AI and Embabel, you can define AI agents by supplying them instructions (prompts), specifying the model (OpenAI or otherwise), listing tools, and composing increasingly sophisticated agent behaviors.
But a good AI Agents programming model is not enough. These agents are ultimately wildly distributed systems and are plagued with all of the problems such systems bring.
- How can they persevere through flaky networks?
- How can they function when LLMs are rate limited?
- How can they run for long periods of time (hours, days, weeks, months) when infrastructure is rarely stable that long?
- How can they effectively blend human agents with robots?
In this workshop, we’ll show you how. Temporal is an open source (MIT license) durable execution framework that brings resilience to AI agents, and in this workshop we’ll show you how it’s done using Spring AI and Embabel. Spoiler: the heavy lifting has already been done for you - Temporal integrates naturally with Spring-based applications, making it straightforward to add durable execution, retries, and long-running orchestration to agentic systems.
Oh, and this approach is not theoretical. Temporal is used in production to make complex, distributed AI-driven systems reliable and resilient — including systems that interact heavily with LLMs and other non-deterministic components.