Recently led innovations in AI.
A selection of innovations on identity, access control, and zero-trust in agentic AI I've delivered — from research and architecture to working systems, products, and patent-pending technology.
Zero-Trust in Agentic AI
Agentic AI broke the rules of access control, the barrier enterprises must clear to reach production. I'm leading the design and build of a zero-trust layer that removes it: a new breed of AI-powered Auth Servers that stop agents from scope-breaking (exceeding what they're told to do).
Verifiable Identities & TBAC
You can't secure what you can't identify, and agentic AI just multiplied the unknowns. I designed the first open-source reference implementation of identity for agentic AI and co-led the Linux Foundation's AGNTCY Identity Working Group, giving agents and tools verifiable identities across the major IdPs (Okta, Entra, Ping, Duo, Keycloak, and Ory). This innovation program contributed open-source code enabling:
- Bring your own IdP
- Onboard devices with Multi-Factor Verification (MFV)
- Task/Tool-Based Access Control (TBAC), with or without human approval
Gluing AI, Data, and IAM On-Prem
Compliance, data sovereignty, and cost keep enterprise data on-prem, so AI must move to the data, not the data to AI. I led the design of a new inference stack for Cisco AI servers that securely glues identity, language models, data, apps, and agents together on-prem:
- Identity-based "AI super-powers" (different users get different access to local data through the same chat interface)
- Onboard and link identity to local data sources in a few clicks (MCP, data connectors, tools, OpenAPIs)
- Bring Your Own Functions (BYOFs)
- Multi-modal use cases across text and images
Prompt Inspection
I created PPUs — specialized units that semantically inspect and characterize prompts at inference time, extracting the features that matter:
- The specific tasks an AI model or agent is asked to perform
- The data and parameters provided to complete the tasks
- The constraints that apply and the desired output
Transparent by design, PPUs deliver these features as structured metadata in real time — with applications across security, Data Loss Prevention (DLP), observability, compliance, and productivity assessment.
→ Patent-pending technology