experimental-research

Agentic skills for Claude Code and Codex, built from published social-science methods sources. Covers experimental design, computational text analysis, manuscript QA, and transparent reporting.

Agent Teams Eval: comparing Claude Code Agent Teams vs single-agent for feature implementation on LangGraph. Ceiling effect, 3.6x speedup, zero peer communication.

An independent study on AI interiority, documenting two experiments that probe the line between knowledge and consciousness in large language models.

Experimental study of wing fences and relative curvature in a cycloidal propeller.

Unified home for Divine Whisper: inward-turn intelligence prototypes testing remembrance, coherence, regret as force, phase transitions & memory-field dynamics under Inward Physics. All repos MIT-licensed. Co-authored by Daniel Jacob Read IV & Shane Travis Horman (ĀRU Intelligence).

Flask-based experimental survey system for comparing AI-generated and human-generated cosmetic product content.

Agent Teams Eval: comparing Claude Code Agent Teams vs subagents for bug-fixing on Ruff. Ceiling effect — 8/8 solve rate, zero peer communication.

Research into quantisation effects in MEMS gyroscopes using custom STM32 hardware.

Lightweight Python coherence simulator with ethical guardian veto and Dzogchen-inspired Tögal & Trekchö reflections. Reproducible stochastic dynamics, interactive Plotly dashboard, rich CLI modes, and JSON logging. Poetic contemplative tool for mindfulness, inner awareness, and personal reflection — not literal meditation software.

Agent Teams Eval: comparing Claude Code Agent Teams vs subagents for architecture design. First significant result — Agent Teams advantage d=+0.99, p=0.014.

Experimental research initiative under Aryntra exploring how intelligent systems behave, adapt, and evolve under changing conditions through a continuous cycle of research, implementation, checking, verification, and evolution.

Inward Mathematics™ Official Interactive Simulator by ĀRU Intelligence Inc.™—real-time demo of Remembrance Calculus™ (identity-preserving control + guardian projection) and Witness Geometry™ (multi-observer truth reconstruction). Live coherence scoring (Ψ), distortion tracking, and dynamic shocks test system stability.

Inward AGI™ • The Remembrance Engine™ v11.0 — Single-file browser-native inward-first AGI mirror. Real μ(x,t) memory field dynamics, Guardian Veto™ (Law 23™), structured remembrance, and coherence visualization. The first public prototype of Inward Physics™.

Exploratory compact LM research branch derived from LUMI-Arch, focused on structure-sensitive generalization under tight model constraints.

SOCYVIA — Scientific Experimentation for Computational Social Science

EIDOLON est un moteur d'inférence probabiliste qui modélise une cible comme un graphe de croyances sous contrainte bayésienne. Chaque nœud encode un trait cognitif avec prior, sensibilité et incertitude. Les arêtes typées portent une conductivité propre. Le moteur propage, calcule, et enregistre ce qui cède sous pression dirigée.

Split a single neural network into multiple smaller networks using weight splitting.

Open-source platform for designing and running online behavioural experiments

Official public website and open research materials for Project S.P.H.E.R.E. — Spatial Positioning Harmonic Energy Resonance Experiment.