Limited edition archival prints available Browse the collection →
Projects
AI governance tooling, security infrastructure, and software products — built at the intersection of product management and engineering.
Launched Products
AI agent governance, safety, and audit
Enforces security policies on autonomous AI agents across frameworks like Claude, GPT, and LangChain. Policy-based controls, human-in-the-loop approvals, semantic threat detection, and cryptographic audit trails.
The fairest way to meet in person
Make it easier for people to show up for each other. Group meetup coordinator that eliminates the "where should we meet?" friction — finds the fairest location using algorithmic geospatial fairness scoring, not the loudest voice or the organizer's neighborhood.
Hospitality reputation management
AI-powered guest engagement and review management for hotels and properties. Audits online presence across booking platforms, voice-matched AI responses, active monitoring, and critical feedback recovery.
Architect the ultimate local LLM hardware
The neutral tool that answers the questions that matter for local AI inference: will this model run on this hardware, how fast, and where is it cheapest? Rigorous compatibility engine, tok/s ratings, cloud vs. ownership cost comparison.
Developer Tooling & Open Source
Durable multi-worker coding orchestrator for Codex, Claude Code, Grok, and shell. One Node runner owns decomposition, worker routing, process supervision, isolated Git worktrees, adaptive review, transactional integration, and recovery.
Scan, score, convert, and coach AI agent configurations across 17 platforms — Claude Code, Codex, Cursor, GitHub Copilot, Gemini, Windsurf, and more. Security scoring via 80+ rules, cross-platform conversion with fidelity reporting, and AI Coach.
Static security scanner for AI CLI configurations. Detects prompt injections, credential leaks, and threats in Claude Code, Codex, Gemini, Droid, Opencode, and other LLM CLI tool configs.
Experiments