I build deterministic governance infrastructure for AI systems.

Phionyx treats large language model outputs as noisy cognitive measurements rather than final answers. The goal is to place a verifiable governance runtime between AI systems and end users: safety gates, ethics gates, telemetry, evaluation standards, state evolution, and audit-first control.

• Phionyx Core SDK — deterministic AI governance runtime
• HearthOS — bounded-authority household AI: the operating principle from Volume I of the Governance Trilogy, demonstrated end-to-end in three browser-only modules (Diagnostic, Weekly Reset, Boundary Script) backed by an open-source TypeScript reference implementation and a free printable Starter Kit
• Phionyx Evaluation Standard — behavioural reliability, safety, coherence, determinism, and long-term stability evaluation
• Governance Node Architecture — multi-gate AI control and release model
• Trace / Wheel & Balance — educational and narrative ecosystem for resilience, decision-making, and non-violent RPG-based learning

• LLM output is not truth; it is a signal requiring governance.
• AI systems need runtime control, not only prompt-level safety.
• Safety, coherence, and telemetry should be structured before response release.
• Evaluation must include behavioural stability, not only benchmark performance.
• Human-facing AI should be explainable, auditable, and interruptible.

• phionyx-research — deterministic AI runtime governance for LLM systems (Python; PyPI: phionyx-core)
• phionyx-evaluation-standard — vendor-independent evaluation standard for agentic AI runtimes
• hearthos — bounded-authority household AI orchestration; TypeScript reference implementation, browser-only demo, Starter Kit PDF (AGPL-3.0)

• Website: https://phionyx.ai
• HearthOS demo: https://phionyx.ai/hearthos
• Substack: https://phionyxresearch.substack.com
• X: https://x.com/phionyx_ai