19D Army Veteran building automated intelligence pipelines, live analytical dashboards, and civic technology tools. First line of Python: October 2025.
🌐 Portfolio → · 📊 Live Dashboard → · 🤖 OSINT ChatBot → · 📜 Bill Translator →
The Regulated Friction Project (v11.3) — Research repository documenting temporal correlations between geopolitical friction events and institutional capital flows. 8 years of data, 66 verified event pairs, 16 independent robustness tests (Opus 4.6). Includes a structured AI context index tracking 14 active leverage nodes and reproducible correlation scripts.
The live dashboard at regulatedfriction.me runs a seven-stage automated pipeline twice daily via GitHub Actions — from Brave Search + Claude node tracking through dual-engine entity extraction (GLiNER2 + Llama Scout 17B), convergence detection, daily intelligence, fact verification, and rhetoric vs. reality gap analysis. Nine-tab dashboard with Chart.js visualizations, interactive entity relationship graph (NetworkX + PyVis), public prediction tracker, and an embedded Gradient AI chatbot.
| Finding | Value |
|---|---|
| Core correlation | r = +0.6196 (p = 0.0004) |
| Convergence correlation | r = +0.6685 (p < 0.0001, n = 229 weeks) |
| Robustness | Permutation (p < 0.001), Granger causality (F = 32.49, p < 0.0001), bootstrap (p = 0.008) |
| Independent verification | 16 statistical tests by Opus 4.6 — all passed |
| Predictions | Falsifiable predictions tracked publicly — failures documented alongside confirmations |
OSINT ChatBot — BYOK Flask chatbot on Render. Users supply their own Anthropic API key (never stored server-side). Knowledge base injected from _AI_CONTEXT_INDEX/ at startup. Rate limiting, CSRF protection, DOMPurify sanitization, full security headers.
Bill Translator — Flask app that rewrites dense legal text to 8th-grade readability while preserving legal intent. Built for Arkansas Act 602. Meaning drift detection, three translation modes, automated Flesch-Kincaid scoring, 31 tests. Deployed on Render.
State Policy Analysis (v1.4) — Tracks 42 federal funding withholding events across 27 states by data center policy posture. Six statistical tests (ANOVA, Kruskal-Wallis, DoWhy causal inference), transparent corrections of prior findings.
UVB-76 Signal Analysis — 15-year empirical analysis of Russia's UVB-76 shortwave station. K-means clustering and permutation testing (p < .0005) identified four operational modes and a shift from training-cycle to event-driven signaling.
- Automated OSINT pipelines — multi-stage LLM extraction, entity disambiguation, convergence detection, rhetoric vs. reality analysis, web-grounded fact verification
- Multi-AI verification systems — cross-model validation (Claude, Brave Search, GLiNER2, Llama Scout, Gemini, Grok) to catch hallucinations before publication
- Live dashboards — serverless HTML/JS dashboards with Chart.js and interactive entity graphs (NetworkX + PyVis), fed by scheduled GitHub Actions pipelines
- Temporal correlation frameworks — lag analysis, permutation testing, Granger causality, DoWhy causal inference
- Civic technology — NLP-powered readability tools, BYOK chatbots, legislative transparency apps
- Reproducible research — every finding ships with verification scripts; failed predictions documented publicly
Build your own live intelligence dashboard in under 15 minutes.
The same pipeline architecture behind regulatedfriction.me — packaged as a turnkey .tar.xz you can deploy to any GitHub repo. Define your own topics, add two API keys, and it runs itself. Twice a day, every day.
What you get:
- Ten-stage automated pipeline — node tracking, dual-engine entity extraction (GLiNER2 + Llama Scout), convergence detection, three-pass daily intelligence, web-grounded fact verification, rhetoric vs. reality analysis, interactive entity graph
- Nine-tab live dashboard on GitHub Pages — no servers, no Docker, no databases
- Self-verifying output — the pipeline fact-checks itself against live web evidence before publishing
- Prediction tracking with full audit trail — confirmed and failed forecasts documented transparently
- Fully configurable — swap every tracked topic, signal, and entity rule from a single JSON config
$50 one-time purchase · Under $10/month to operate · GitHub Actions free tier
Buy on Creem → · Buy on Gumroad →
See the Product Description for the full pipeline architecture walkthrough.
- Threat intelligence and OSINT analyst roles
- Investigative data support and newsroom partnerships
- Freelance intelligence pipeline development and consulting
- Open-source collaboration on accountability tools
Contact: Discord and email in profile description
Last Updated: April 1, 2026