For engineers managing AI agents in production
Summary
An engineering discussion exploring how to change an AI agent's authority and permissions after it's already deployed in production, comparing approaches like config deploys, IAM/RBAC, policy engines such as Cedar and OPA, and gateway/proxy layers.
Similar Articles
Those of you running AI agents in prod — how are you actually managing their permissions?
The article asks how engineers manage permissions for AI agents in production, highlighting common problems with broad access and lack of audit trails.
Should AI agents be allowed to deploy or change production resources directly?
A discussion on whether AI agents should be allowed to directly deploy or change production resources, or if human approval should always be required, particularly for small teams.
How to prevent AI agents from taking unintended or harmful actions in production
A developer discusses challenges in deploying AI agents to production without causing unintended harm, seeking advice on control mechanisms like least privilege, shadow mode, rate limits, and approval workflows.
People running agents in production: how do you control what they're actually allowed to do?
A developer seeks advice on how to control and bound AI agents' actions in production environments, particularly when they interact with real systems like databases and customer data, asking about current practices and whether this is a known headache.
How are you handling authorization for AI agent tool calls in production?
The article discusses challenges in handling authorization for AI agent tool calls in production, including issues like context drift and multi-turn inconsistencies, and requests community input on effective architectural patterns.