Make your own event-sourced agent harness using stream processors — Jonas Templestein, Iterate

The abstraction is three things: state, a synchronous reducer that derives state from events, and an after-append hook for side effects. The split matters: when your program restarts after 100 events, you want to catch up state without replaying LLM requests. Everything that happens (streaming chunks, tool calls, errors, circuit breaker triggers) is an event in the log. The interesting part is deployment. Jonas demos "dynamic worker configured," an event whose payload is a JavaScript string containing a processor. Append it to any stream and that stream becomes an AI agent without server or dependencies. The broader implication: processors from different authors on different servers can compose against the same stream, and a safety checker can inject context in a 200ms window before an LLM request without blocking the agent if it doesn't make it. Speaker info: https://x.com/jonas   / jonashuckestein   https://github.com/jonastemplestein 0:14 Introduction and workshop overview 1:33 Concept of event-sourced agent harnesses 2:24 Desired characteristics: Extensibility and composability 3:14 Agents on the edge and public routability 4:20 The distributed nature of agents and potential pitfalls 5:40 Introduction to the event stream architecture 8:38 Working with the API via curl 10:48 Handling events and error states 11:57 Circuit breakers and stream management 13:49 Scheduling tasks and event subscriptions 15:17 Q&A: Architectural philosophy 19:17 Demonstrating the SDK and TypeScript integration 30:06 Defining stream processors and reducers 47:34 How stream processors function in production 50:35 Dynamic workers and deploying via event payloads 58:45 Discussion on before-hooks versus eventual consistency 1:02:16 Future outlook and wrapping up