Access reviews are one of those processes that exist at the intersection of security and frustration. They’re necessary, they’re mandated, and they’re often deeply manual. At Lumos, we asked: what if an agent could do the heavy lifting?
This project explored how AI-driven insights could surface the right context at the right moment — flagging anomalies, surfacing usage patterns, and reducing the cognitive load on reviewers so they could focus on decisions rather than data gathering.