Become a sponsor to Justin K. Lietz
Support Zero-Trust Physics & Cognition Research
I build "zero-trust" auditable physics and cognition systems that can be proven wrong with clear instructions on how to do so.
The Void Dynamics Model (VDM) is an independent research program with two tracks: a background-independent metriplectic field theory, and a zero-training cognitive runtime built on sparse, event-driven neural graphs. We treat experiments like software releases—predictions are preregistered, mathematically gated, and published with contradiction reports by default. No narrative padding, just falsifiable code.
Why Sponsor?
Independent research requires independent compute. By bypassing traditional academic funding constraints, we can execute high-leverage, zero-to-one engineering without bureaucratic friction.
Your sponsorship directly accelerates this architecture. Funds are channeled into scaling enterprise compute clusters, which allows us to run heavier validation sweeps, stress-test the cognitive runtime, and push the limits of our emergent neural dynamics.
What You Get
Sponsors gain exclusive, behind-the-scenes access to the machinery that powers VDM. By supporting this work, you unlock:
- Adversarial Tooling: Access to private Jupyter notebooks, SymPy scripts, and rigorous Lean 4 packages explicitly designed to test and attack our architectural claims.
- Early Access: Unreleased code, experimental branches, and early builds of the zero-training cognitive runtime before they hit the public repositories.
- Raw Data & Insights: Inside access to validation sweep results, architecture discussions, and the direct engineering logs of Neuroca.
- Direct Collaboration: Priority review of your own reruns, contradiction reports, or attempts to falsify our findings.
If you value reproducibility, strict pass/fail thresholds, and making the ability to fail quickly an advantage—help stress-test the frontier.
I sold my $13K AI workstation and furniture to build the Void Dynamics Model solo from an empty apartment. $2K/mo (Current): Secures my survival runway for 100% full-time focus. $5K/mo: Rebuilds my compute arsenal (AMD accelerators) for heavy validation sweeps. $10K/mo: Falsification bounty for researchers who break our math.
Featured work
-
justinlietz93/cortex-engine
Phase Calculus world simulation engine for physics, video games, AI models, operating systems, robotics, and more.
HTML 2 -
justinlietz93/vdm_rt
Real-time, zero-training, self-managing topological graph model with emergent learning driven by sophisticated, rich spiking neural dynamics. Original core runtime engine for the Void Dynamics Model.
Python -
justinlietz93/orchestra
Three agent project tracking desktop application for recording Operator, Guardian, and Auditor interactions without losing the project line, source artifacts, or handoff state.
Python -
justinlietz93/TeXterity
A native, offline-first desktop game written in Rust for learning to type LaTeX used in graduate mathematics and physics.
Rust 1 -
justinlietz93/Taoism-Mathematics-Research-Project-Internal
A collection of research artifacts exploring Phase Calculus and connections to Taoist texts and related mathematics.
Jupyter Notebook -
Neuroca-Inc/Cogito-Research
Cogito: Computational framework employing a Syncretic Catalyst (synthesis) & rigorous Critique System (analysis) autonomously. Deeply accelerates your expert research on scientific literature via L…
Python 2