AI Agents Conference · November 23–27, 2026
Rijwan Ansari
Tech Lead/Consultant - Enterprise CRM
Session
Who Watches the Coding Agents? Building an AI Architecture Guardian for Autonomous Software Engineering
AI coding agents are rapidly moving beyond autocomplete and code generation. Today, an agent can explore a repository, understand requirements, generate code, create APIs, modify infrastructure, run tests, diagnose failures, open pull requests, and even orchestrate deployments. But there is a problem we haven't solved yet: Who watches the coding agents? When developers were the primary builders, organizations created architecture reviews, security reviews, coding standards, ADRs, governance boards, and engineering policies around them. But what happens when the developer delegates more and more of the engineering work to autonomous agents? Who checks whether the agent introduced the wrong dependency? Who notices that a new service violates the target architecture? Who detects that sensitive data is being sent to an AI service? Who knows whether the infrastructure created by an agent follows enterprise standards? And perhaps most importantly: Can another agent continuously review the architecture being created by coding agents? In this session, we will build exactly that. We will explore how to create an AI Architecture Guardian that can inspect a real software repository, discover its architecture, understand the technologies and dependencies being used, identify security and architectural risks, reason against engineering policies, and provide actionable recommendations.
Share your speaker card
Speakers: download this card and post it on LinkedIn, X or Instagram. Links to this page preview with the same design.
Full conference agenda
See every session across the AI Agents Conference November 2026 line-up.