Products built to work.

An in development MMA promotion management game for iOS and Android. Players build a roster, negotiate contracts and broadcast deals, manage finances, assemble fight cards, and navigate the business behind each event.

Problem

A mobile management game needs its roster, contracts, finances, events, and AI driven characters to work together without sacrificing predictable game rules.

Solution

A shared Java/libGDX simulation handles the promotion and fight systems, while a Python and Flask bridge supplies bounded AI behavior for negotiations, matchmaking, media reactions, and character stories.

Why I built it

Justin built MMA Promoter to explore the strategy of running a fight promotion while testing how bounded AI could add variety to characters and negotiations.

A source grounded dataset generation system with explicit model contracts, deterministic Python validation, a trained critic layer, persistent job state, and operational evaluation gates.

Problem

Turning raw documents into dependable training data requires more than an unconstrained prompt: source claims, schemas, metadata, formatting, and duplicate handling all need verifiable controls.

Solution

The documented pipeline separates extraction, planning, bounded generation, deterministic validation, second pass critique, and packaging, with progress polling and observability across the workflow.

Why I built it

Dataset Forge explores how a simple dataset building interface can sit on top of a rigorous asynchronous workflow without hiding the reliability work behind it.

A revenue recovery system that helps small and midsize businesses respond to customer calls and inquiries, continue conversations, send follow ups, and accept online orders through an enterprise mobile app.

Problem

Small and midsize businesses can lose valuable revenue when customer calls or inquiries go unanswered. A single missed customer can represent hundreds or thousands of dollars in lost business.

Solution

VeloBook responds to inquiries, talks with customers, sends follow ups, and provides a mobile enterprise app where the business can manage conversations and take online orders.

Why I built it

Justin created VeloBook to help businesses recover revenue that would otherwise be lost when they cannot respond to every customer immediately.

A protected content-planning and drafting workspace that helps turn an idea into an editable LinkedIn post while keeping review and publishing under the visitor’s control.

Problem

People want to share their ideas and expertise on LinkedIn and create meaningful outreach, but planning, researching, and writing consistently takes more time than many have.

Solution

The protected workspace turns a content brief into an editable draft, adds a review checklist, and supports copying or exporting the result without sending anything to LinkedIn.

Why I built it

Justin built the LinkedIn Content Engine to automate the repetitive parts of content creation and give people an AI assistant that helps turn their thinking into posts while keeping them in control.