# Let's integrate AI Agents in Event-Sourced Systems — Divakar Kumar, FlyersSoft

Source: [AI Engineer](https://www.youtube.com/watch?v=o6U_2vd967Y)  
Feed7 permalink: https://feed7.dev/p/let-s-integrate-ai-agents-in-event-sourced-systems-divakar-kumar-flyerss-1w3t01t  
Published: 2026-07-30T00:00:06.000Z  
Trust: Source Linked (source_linked)

## Why Included

Add agents beside deterministic fraud systems, not in place of them: project cross-service events into a semantic layer, send only uncertain cases to specialist agents, then emit the verdict back into the saga.

## Source Summary

The design keeps rules and traditional ML as **tier one** for routine transactions, routing gray-zone cases to **tier two**. Events from transaction, account, device, and payment contexts feed a semantic projection used by risk and behavior agents; a third agent produces the verdict.

## Practical Implication

For an existing event-driven system, build agent context from read-optimized projections instead of querying raw event stores or collapsing bounded contexts. Return the result as another event so the agentic step participates in the established saga.

## Agent-Ready Context

The design keeps rules and traditional ML as **tier one** for routine transactions, routing gray-zone cases to **tier two**. Events from transaction, account, device, and payment contexts feed a semantic projection used by risk and behavior agents; a third agent produces the verdict.

For an existing event-driven system, build agent context from read-optimized projections instead of querying raw event stores or collapsing bounded contexts. Return the result as another event so the agentic step participates in the established saga.

The demonstration used synthetic data and reported no fraud-accuracy, latency, or cost results. A third agent does not make a verdict deterministic, and the orchestration loop needs an explicit stopping condition to avoid running indefinitely.

## Connected Context

Feed7 judgment across 297 accumulated Signals:

This adapts agent orchestration to an existing event-sourced architecture instead of replacing it: deterministic rules handle routine cases, semantic projections provide bounded context for gray zones, and verdicts re-enter the saga as events. It reinforces staged automation and minimum-path routing, but the fixed multi-agent chain conflicts with evidence that specialist pipelines can lose coherence; synthetic demonstration and absent stopping rules leave reliability unproven.

- [AI tools for Forward Deployed Engineering — Vasuman Moza, Varick Agents](https://feed7.dev/p/ai-tools-for-forward-deployed-engineering-vasuman-moza-varick-agents-12kjg79) — Its risk-based automation boundaries reinforce keeping routine transactions deterministic and sending only ambiguous cases to agents.
- [Why We Killed Our Multi-Agent Pipeline — Subbiah Sethuraman and Abhilash Asokan, ZS Associates](https://feed7.dev/p/why-we-killed-our-multi-agent-pipeline-subbiah-sethuraman-and-abhilash-a-0fmz3z3) — Its failed specialist chain warns that separate risk, behavior, and verdict agents may lose context; bounded subagents with one end-to-end owner may be safer.
- [Do AI Agents Know When a Task Is Simple? Toward Complexity-Aware Reasoning and Execution](https://feed7.dev/p/2607-13034v1-03g7ghx) — Complexity-aware minimum-path execution supports the two-tier design: use the simplest verified route first and expand to agent reasoning only when needed.
- [How Forward Deployed Engineering is done at Kepler — Vinoo Ganesh](https://feed7.dev/p/how-forward-deployed-engineering-is-done-at-kepler-vinoo-ganesh-0thtvuc) — Its smallest-useful-fix discipline argues for validating one narrow event-driven intervention before expanding the proposed architecture across bounded contexts.

## Context Map

- Layer: agent
- Domains: data
- Topics: multi-agent, context-engineering, harness-engineering

## Uncertainty

- The demonstration used synthetic data and reported no fraud-accuracy, latency, or cost results. A third agent does not make a verdict deterministic, and the orchestration loop needs an explicit stopping condition to avoid running indefinitely.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.