Portfolio

Research

Original systems, protocols and architectural ideas

Whitepapers, open architecture and reference implementations, and research writing on AI-agent execution, identity, security and distributed coordination.

Research areas

AI Agent Execution

Governance models for controlling agent actions, permissions, evidence and settlement within individual systems.

Identity & Security Architecture

Protocols for agent identity, authorization, accountability and verifiable interaction across system boundaries.

Distributed Coordination

Architectures for coordinating objectives, commitments and outcomes between independently governed systems.

Featured research

Whitepaper · Published · Version 1.2

Outcome Coordination for AI Agents

A protocol architecture for governing execution within systems while coordinating commitments across trust boundaries.

Explore the research

Published research and open systems

How this work is published, specified and demonstrated.

Whitepapers

Formal, versioned and citable technical publications covering AI-agent execution, security and distributed coordination.

View whitepapers

Lattice Capability Runtime

An open-source runtime for structured AI-agent execution through explicit capabilities, bounded authority and controlled interaction with enterprise systems.

Architecture diagrams · Source code · Documentation

Covenant Layer Protocol

An open protocol for coordinating outcomes across trust boundaries through objectives, offers, commitments, evidence and settlement.

Protocol diagrams · Source code · Technical specification

RealTimeAuth (RTA)

A low-latency continuous authorization protocol re-evaluating token permissions at the execution boundary during active agent sessions.

Whitepaper · Connection-bound AuthZ

agrep

A deterministic query-planning retrieval engine with domain-aware ranking and JSONL output for autonomous agent workflows.

Query planner · Ranking · JSONL schema

Current research direction

My current work examines how autonomous systems can coordinate outcomes without requiring shared ownership, centralized execution or unrestricted trust. The focus is on explicit objectives, bounded authority, commitments, evidence and settlement. This work is developed through the Lattice Capability Runtime and Covenant Layer Protocol, with architecture diagrams, specifications and reference implementations published openly.