Welcome to my virtual soapbox. This site contains my unique way of expressing my creativity through code, UX, and words. It showcases my Github and AI projects and provides a space to publish my thoughts.
I’m a software engineer with diverse experience architecting and building robust solutions that drive business growth. My expertise lies in tackling complex logic and delivering high-impact features. I’ve worked for startups of various sizes and large corporations delivering features at a global scale. I thrive on ownership, transforming challenging requirements and legacy systems into clean, maintainable code that directly contributes to business value and user satisfaction.
Orphic DJ is an interactive album of original generative music you shape to your taste, playing live on your device, endlessly. Press play and no two listens are ever the same. No streaming, no library, no playlists.
Inside you get 11 vibes, 11 visualizations, 8 generative tracks, 2 turntables, and infinite music that never repeats.
Four knobs capture the essence of any track. Energy moves from calm ambient drift to a driving beat. Complexity goes sparse to layered. Mood runs bright to dark. Space opens from intimate to vast.
Open the DJ module and jam. Two decks and a crossfader set the stage, then you embellish the mix live. Trigger drop effects on the fly, push the delay into runaway feedback loops, and scratch across the beat. Filters, stutters, freezes, brakes, and ring mod are all a tap away.
Orphic DJ is not a one-off. It is assembled from the same reusable feature set that powers Orphic-FM: a shared generative engine, a C++ DSP graph, and a catalog of self-contained feature modules. It’s designed so an instrument can be lifted out and shipped on its own, and Orphic DJ is the first to make that leap. The same engine runs across Android and desktop, stays completely offline with no account or ads, and keeps a home-screen widget a glance away.
Available free on Google Play. Source on GitHub.
Keeping up with new Android libraries and trends is always a challenge. One technique I use is to always have a reference app handy. These apps showcase the latest Android architecture and patterns for UI and business logic separation, navigation UX and theming, data retrieval, threading, dependency injection, and testing.
The latest incarnation is called Recipes. This app conflates Meal and Code Recipes, creating a UX to demonstrate the techniques it describes. The goal was to brush up on the latest Android tech stack and pack as many features as I could into an app created in my “spare” time.
One data source is from TheMealDB which has categorized data with textual and visual information describing a semi-complex process (aka Meal Recipe). The app features category and cuisine browsing, ingredient lists with measurements, and step-by-step cooking instructions.
The UI showcases Material 3 design with adaptive layouts for phones, tablets, and foldables. Shared element transitions create fluid navigation between views.
The other data source was my 🧠, describing the code used to build the app in the form of Code Recipes. These recipes describe the wiring and plumbing used to create a visually compelling Android Application.
Topics include Metro dependency injection, Compose navigation with Nav3, shared element transitions, and adaptive layouts. Each recipe includes markdown-formatted descriptions with code syntax highlighting.
Recipe Maestro is the in-app chatbot that answers questions about meal and coding recipes. The agent configuration provides randomized mood prompts for more intesting engagement.
The agent was built using Koog and integrates with the Gemini models. Multiple tools were created to provide access to remote and local data sources and allow for AI recipe creation and navigation. The Agent Screen also displays token and tool usage metrics.
🤔đźŽđźŽ¶ Mood Music is a game where users guess the mood of the music they hear. Players are presented with colorful buttons, each assigned a secret mood. Press a button to hear AI-generated music, then guess which emotion it evokes.
The game features multiple difficulty modes. Easy mode shows only the moods used in that round. Hard mode tests your emotional IQ by showing all possible moods.
When you press a button, the AI generates unique music based on the button’s hidden mood. The mood selector opens with emoji-labeled options like 🎉 Happy, 🌧️ Sad, or ⚡️ Energetic.
Each button has a randomly assigned icon with its own distinctive color palette. The music stops when you make your selection, building anticipation for the reveal.
Once all buttons have guesses, the Reveal Button activates. Correct guesses light up green, incorrect ones flash red. Hit the Victory button to hear an AI-generated celebration song.
Built on Kotlin Multiplatform with Compose Multiplatform UI, the app uses Lyria Realtime for AI music generation. Each mood has carefully crafted prompts—from “intense Heavy Metal” to “calm ambient textures.”
I’ve always searched for the elusive write once, run everywhere solution. Kotlin is my favorite language and I enjoy Coroutines and Flows. All the new build tools and libraries have me excited about what comes next.
This repo contains my first KMP app, Koin app, Game Loop, All Compose app, and Vibe Code project. The app runs on Android, iOS, Desktop, and Web with shared code.
When Junie was released, I built an app that compiled and ran on all KMP platforms with an almost working game. It reminded me of how I start features—crank out code until something works, then stash and start making commits.
The Stars rendering logic showed how to use the Canvas Composable. The terrain generation laid out landing pads with mountains, craters, and rocks. The PhysicsEngine math worked out well.
The project leverages Kotlin Multiplatform with Compose Multiplatform to deliver on the “write once, run everywhere” promise. This runs on Android, iOS, Desktop, and Web with shared code. There’s a clean MVVM separation between UI, ViewModels, and data models.
Game logic is separated into: Physics engine, Camera system with dynamic zooming, procedural terrain generation, and sound service abstraction.
This has been the most fun coding since the last canvas draw feature I worked on. I produced a working game using new technologies in a little over a fortnight. The highlight was producing code I could learn by example from.
Learning to communicate with an LLM will be a fun challenge. The skills of asking the right questions, validating results, and iterating are still critical while using AI.
Remote
Led a distributed engineering team through the full development lifecycle of the official NHL app for Android Mobile, Fire Tablet, Android TV, and Fire TV platforms.
San Francisco, CA
Beaverton, OR (WHQ)
Portland, OR