From guardrails to fixes: Enforcing policy with open remediation language and deterministic AI

Most organizations stop at detection. Scanners generate findings. Policies define what good looks like. But the actual fix still depends on manual effort, context switching, and engineering time. This workshop focuses on closing that gap. The speakers will introduce the Open Remediation Language, a new specification designed to convert guardrails into enforceable, reusable remediation logic. They will explain why a new specification was necessary, what problem it solves, and how it enables consistent, validated fixes across repositories.

Through practical use cases, they will demonstrate:
• Converting Sentinel- or OPA-style policies into remediation logic
• Automatically fixing static analysis findings while maintaining control
• Applying deterministic fixes inside IDE workflows
• Bounding generative AI outputs with validation and verification
• Selectively accepting or rejecting remediations before commit

They will also reference real-world implementation patterns inspired by large-scale environments, including high-volume repository remediation and policy enforcement at scale. Attendees will leave with a clear understanding of how to:
• Burn down IaC technical debt
• Implement guardrails without slowing delivery
• Connect platform engineering, security, and operations through validated remediation workflows

This session is ideal for platform engineers, DevOps leaders, and security architects who want to move beyond alerts and make policy enforcement practical and measurable.