This page contains published projects focused on artificial intelligence, cybersecurity, automation, and software engineering.
The work is centered on practical implementation. Each project is designed to explore a specific technical problem, produce a usable capability, and make the underlying approach available for others to review, test, adapt, and extend.
Projects include agentic AI systems, cybersecurity platforms, engineering tools, and technical learning environments. The common focus is building working software that can be applied to real problems.
-
ATO Bot
Source Code: GitHub Repository Technical White Paper: ATOBot Technical White Paper Installation Tutorial: Watch on YouTube Status: Active Development ATO Bot is an open source platform for performing NIST SP 800 53 assessment and supporting the broader federal Authority to Operate process. I built it to see how far AI could be applied to authorization…
-
BuildLLM Learning Lab
BuildLLM teaches you how a decoder-only language model is built, trained, evaluated, and scaled. You will work with the same implementation used by all three model iterations in this repository. The smaller models are not separate tutorial programs. Architecture, runtime, and data choices change through configuration and command-line arguments while the training framework remains the…
-
Candidate Selection Board
The Candidate Selection Board is a self-hosted platform that utilizes a 13-agent AI model to facilitate structured hiring decisions through expert evaluations and consensus recommendations. It includes multi-stage reviews, robust audit trails, and FISMA Moderate compliance for federal use, ensuring transparency and defensibility in the hiring process.
-
Nutrition AI Pro
Summary Nutrition AI Pro is an agentic AI recipe and nutrition-content production platform for fitness-focused teams. It helps content teams move beyond one-shot prompting by combining structured recipe generation, iterative quality critique, human review, content conversion, prompt management, and operational controls in one self-hosted application. Use it to produce recipe cards, blog posts, social assets,…
-
CyberTabletop
Summary CyberTabletop is an agentic AI cybersecurity tabletop platform for running live, scored incident response exercises. It helps security teams move beyond static slide decks by combining AI-generated scenarios, adaptive injects, role-based player decisions, facilitator controls, real-time scoring, and structured debriefs. Use it to rehearse ransomware, business email compromise, data breach, insider threat, DDoS, supply-chain,…