Tym Huseby
A preregistered, independently audited GHOST study: ecological selection on frozen standing heritable variation enriched genotype distributions with greater inherited capacity for causal, durable, adaptive lived individuality through a preexisting HabitLearner—without inventing the learner, and without claims of consciousness or general intelligence.
Tym Huseby
GHOST studies whether a living digital ecology can drive foraging competence upward through survival and reproduction alone, with no external fitness grader.
Second GHOST paper: ecological selection enriches inherited capacity for causal lived individuality via HabitLearner plasticity—preregistered and independently audited.
Read noteEmergent ecological selection lifts intrinsic foraging competence from random genomes—no external grader, replicated across independent seeds.
Read note
Apache-2.0 agent memory with benchmarked recall, MCP tools, and self-hosted Docker.
Read note
Self-hosted routing proxy with task classification, fallback chains, and a live dashboard—MIT-licensed on GitHub.
Read note
v1 bundle, install flow, and local-first workspace packaging.
Read note
Dashboard, voice, memory, plugin, and scheduler work continue to shape the assistant runtime layer.
Read note
Compact instruction format and model routing strategy are being framed around cost control and cleaner orchestration.
Read noteInternal operations hub structure, workflow modules, and shared visibility patterns are the current focus.
Read note
Voice-first personal progression and health-tracking mechanics are being framed around retention, clarity, and long-term use.
Read note