A blueprint for building a generic... (Ashwin Sidhalinganahalli, Fletcher Ramee)
summary
A hub‑and‑spoke control plane called Guard stores all policies and pushes them to Topaz sidecar evaluators that run next to apps for millisecond P99 latency; fallback options include a central evaluator for very fresh or huge policies and a transpiler to platform native rules for mesh and Kubernetes. The deployment handles 1M decisions/sec in a case study, adds short lived caches, requires initial policy sync before app start, and yields single‑pane visibility plus just‑in‑time and orphaned‑permission cleanup.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:41The authorization problem
- 8:21The blueprint solution
- 16:52Authorization architecture
- 21:25Securing AI agents
- 24:37Platform-level authorization
- 27:05Lessons from deployment
- 31:04Driving adoption
- 33:29Key takeaways
- 35:59Recap and next steps
- 36:40Q&A session
related talks
30mThe Next-Gen Elephant in the Room
36mDynamic and Context-Aware Security Policies for the Cloud Using...
36mCentralized Policy Management
30mRedefining Access Control: Scaling Policy as Code for Humans and AI Agents
31mAccess Policies for Hybrid Cloud Environments
17mSecurity and Flexibility: Decouple Policy to Enable Architectural Choice
30mCan security speed development? Network policies, but with people in mind
17mOpen Policy Agent Intro