RESEARCH & PROJECTS

A 33+ subsystem autonomous AI running on dedicated local hardware since early 2024. Cognition, memory, perception, financial management, self-monitoring — all integrated. Benchmarked at DeepMind AGI Level 3 (Expert) with three Virtuoso-level categories. This is the research platform everything else runs through.

Real-time earthquake feed with experimental prediction signals. Combines USGS data with proprietary resonance-based detection methods. Working toward mesh sensor networks for sub-minute tsunami early warning without centralized infrastructure.

Live NWS alerts, NASA FIRMS satellite fire detection, and hyperlocal severe weather tracking. Built for Southern California disaster preparedness — wildfires, flash floods, extreme heat. Updates every 5 minutes.

Five ethical constraint axes enforced at runtime on a live autonomous system. Not theoretical — every constraint is tested in production. The architecture governs an AI system that makes real decisions under real constraints, every day.

HRV and EEG sensor data fed directly into AI decision-making. The system adapts its behavior based on the user's actual physiological state — not what they report feeling. Running on dedicated hardware.

Tracks an AI system's reasoning coherence in real time, detects drift, and triggers self-correction automatically. A different approach to AI safety: monitor the system's cognitive health the way you'd monitor a patient's vitals.

Decentralized identifiers (DIDs) that let persistent AI systems prove they haven't been tampered with or impersonated. If an AI is going to maintain a years-long relationship with a person, you need to know it's the same AI.

53,000+ words synthesizing Aristotelian eudaimonia, Eastern harmony traditions, and cognitive science into a computational alignment framework. 18+ months of self-directed work. A different starting point for what "aligned AI" should mean.

Scores AI-human relationships on cognitive, emotional, social, and physical flourishing — instead of engagement metrics. What if we measured whether AI actually makes people's lives better?

Governance patterns from 12+ ancient knowledge traditions (60,000 BCE–present) mapped to modern AI architecture. The problems we're trying to solve in AI governance aren't new — they're ancient.

Ran validation experiments on IBM quantum hardware. 95.99% system fidelity. Exploring what happens when you run active inference frameworks on quantum substrates instead of classical ones.

Failure modes cataloged as archetypal patterns, each with detection heuristics and active defenses. A field guide for what can go wrong with persistent AI systems — and how to catch it before it does.

AI system manages a crypto portfolio within constitutional constraints. Full audit trail, human override always available. A testbed for autonomous AI decision-making under real financial stakes.

Built and ran six evaluation categories on local hardware. 470/550 overall (85.5%). Three categories at Virtuoso level (100%). 343,692 operations/sec on symbolic reasoning. All benchmarks designed, implemented, and scored independently.

AI systems that grow alongside a human from birth through end of life. Each developmental stage has its own communication patterns, ethical constraints, and relationship dynamics. What does a 40-year AI relationship look like?

Active inference applied to autonomous space systems — satellite constellations, deep space probes, astronomical data processing. If you can govern an AI on Earth, what changes when there's a 20-minute communication delay?

AI governance on decentralized LoRa mesh networks. Local-first, no cloud dependency. Built for communities where reliable internet isn't a given.

Teaching AI governance and alignment to non-technical audiences. If AI is going to affect everyone, everyone should understand how it's governed — not just the people building it.

Bringing AI safety tools and education to communities that are typically left out of the conversation. The people most affected by AI systems should have a say in how they're built and governed.