This is an evolving list of the major projects I work on. Each one represents something I care about — from infrastructure and open protocols to browser-native frameworks, games, and developer education.
I’ll expand this page over time as projects evolve or new ideas take shape.
🛰 Xode RPC
Description
Xode RPC is a public-good infrastructure project — a highly available Arbitrum RPC service that helps keep Livepeer orchestrators online and healthy. I run and maintain a fleet of RPC nodes, handle patching, pruning, upgrades, and provide peace of mind to those who rely on consistent service.
How It’s Used
Several orchestrators in the Livepeer community rely on Xode RPC. Some pay in ETH, others via Stripe in USD, but the goal remains the same: keep orchestrator nodes running reliably and confidently.
Why It Matters
For a long time, RPC outages were a major problem in the Livepeer network. This project helps solve that. It’s a paid service, but it operates like a public good — reducing uncertainty for node operators and raising the overall health of the network.
Future Ideas
- Stay focused on Arbitrum RPC (no plans to expand to other chains for now)
- Continue automating maintenance tasks
- Document reliability metrics and uptime stats
☁️ Livepeer Cloud SPE
Description
The Livepeer Cloud SPE (Special Purpose Entity) is a treasury-funded initiative designed to make Livepeer’s AI and video infrastructure easier to use, experiment with, and build on. I co-founded this project to offer real-time tools, transparent dashboards, and public infrastructure that grows network usage.
What I Maintain
- Free-to-use RTMP + AI inference gateway powered by Livepeer
- Performance dashboards showing orchestrator rewards, fees, and reliability
- Security patches and protocol upgrades for Livepeer nodes
Next Steps
We’re planning a new treasury proposal focused on growing demand. The goal is to use the Ziply Games platform as a real-world demo of Livepeer’s AI video inference capabilities — including funding inference credits and boosting marketing around it.
FIFA World Cup 2026 is our internal target — we want something playable and exciting in time for that global moment.
Future Ideas
- Run real-time AI inference on interactive games
- Onboard more orchestrators into experimental workloads
- Share knowledge and code once unique use cases are proven
🎮 Ziply Games
Description
Ziply Games is a platform for quick, browser-based, multiplayer games that require zero installation. It’s an experiment in SEO, Livepeer integration, and browser-native game development.
Games So Far
- Tic-Tac-Toe
- Rock Paper Scissors
- Hangman
- In-progress: custom games like Scattergories-style word puzzles, crossword generators, and more
User Contributions
While the core games will be curated, users will be able to generate parts of the game experience — like word lists for puzzles or themes for rounds.
Why It Matters
This project is both a playground and a real-world product. It supports the SPE’s mission by creating demand for Livepeer’s AI inference layer, especially as we begin to integrate real-time video and generative content.
Future Ideas
- Monetization via ads or game-based microservices
- Real-time generative avatars and video via Livepeer inference
- SEO as a growth strategy, not an afterthought
- Potential demo app for how local-first, multiplayer, AI-driven browser games can work
🧱 Local-First Framework (Web Components + Streams)
Description
I’m building a lightweight, game-agnostic frontend framework that’s fully browser-native — no React, no Angular, no build step. Just Web Components, IndexedDB, and reactive streams.
Core Features
- CQRS + Event Sourcing patterns
- IndexedDB-backed state with
liveQueryupdates - Automatically managed reactive streams to DOM
- No framework lock-in, no npm bloat
Why It Matters
Frontend development has become unnecessarily complex. This framework is an effort to bring back a “view-source” era of simplicity — while still being reactive, testable, and capable of building serious apps.
Current Status
I’ve gone through several iterations. Early attempts using raw browser events hit limitations — too brittle, too entangled. Moving to CQRS + observable streams has solved many of those headaches.
Future Ideas
- Open source the framework
- Package starter templates for game devs or indie creators
- Offer it as a modern replacement for React in small-to-medium projects
🧠 mikezupper.com + Dev Education
Description
This site is the hub for everything I work on — from Livepeer to games to the web platform itself. It’s where I document, reflect, and eventually teach.
Current Work
- Weekly blog series (“What I Built This Week”)
- Project updates and architecture discussions
- Reflections on tooling, tradeoffs, and how to build lean
Upcoming Content
- Dev newsletter (free + paid tiers)
- Screencasts, walkthroughs, and visual architecture docs
- Short and long-form courses:
- Backend with Rust
- Web development without frameworks
- Local-first architecture and data modeling
Community Ideas
- Launch “The Zoop Troop🫡⚡️” — a perpetual learning community
- “Professor Zoop” persona for storytelling and teaching
- Podcast focused on developer tooling, architecture, and indie building
- Workshops, merch, and digital events
More projects coming soon…
