Nine years across London agencies and Australian enterprise. SSR-first delivery across Next.js, Nuxt, and Astro, test pipelines from unit to E2E, performance-critical government properties, and the component systems and discipline that come from building at institutional scale. The foundation.
Building fast, accessible frontends for products that run at real scale.Building fast, accessible
frontends for products
that run at real scale.
Frontend engineer. Product builder. Based in Brisbane, AU.
Latest, onchain-ui merged into the shadcn registry
A CV tells you where I've worked. It doesn't show you everything between the idea and the release: sprint planning, T-shirt sizing, the ticket that looked simple until you picked it up, and the days spent turning a complex system into an interface that feels obvious. The frustration. The breakthrough. The satisfaction when it all comes together: a versioned build moving cleanly through the release pipeline and out to thousands of users, a page transition that finally feels right, an API response shaped exactly the way it should be.
Before I wrote a line of code, I trained as a chef. It turned out to be less of a leap than you might think. Both demand preparation, precision, a willingness to keep improving, and a clear head when something goes wrong. Those lessons have stayed with me through fifteen years of building software. The tools have changed. The way I approach the work hasn't.
~15 years
Writing production code
4 years
Real-time, high-stakes systems (DeFi)
Enterprise scale
Squiz Matrix, engineered the CMS, not just sites on it
AI engineering
Builds AI products and codes with AI daily, ships faster because of it
Standards-driven
WCAG AAA, design systems, reusable components & craft
Founder
ClawOps built end to end, product decisions, not just tickets
A decade of shipping interfaces people actually use.
Built a commercial SaaS AI assistant platform on top of OpenClaw. A monorepo: React frontend, Convex backend, automated Hetzner VPS provisioning via SSH, Stripe subscriptions, and per-customer OpenRouter API keys with spending caps. The key differentiator is the Companions platform, with installable AI personality layers, persistent memory, their own data stores, API integrations, and visual dashboard UIs.
Co-founded and led engineering on a full product suite built on Next.js with SSR and mixed rendering strategies, built for real-time data and sub-second responsiveness under load, including swap, bridge, staking, yield farming, and an NFT platform, later spun out into a standalone on-chain index protocol in the DeFi space. Shipped across four chains, $1M+ revenue generated, community in the thousands.
Things I've built
and shipped.
ClawOps
SaaS / AICommercial SaaS AI assistant platform built on OpenClaw. Each customer gets a personal Telegram bot on a dedicated Hetzner VPS, live in minutes. The Companions platform is the real differentiator, with installable AI personality layers, domain-specific knowledge, QMD-indexed persistent memory, their own data stores, third-party API integrations, and dedicated visual dashboard UIs. First companion: Entertainment Buddy (alpha), tracking taste profiles, artists, movies, gig history, and SoundCloud sets with smart recommendations.
Highlights
- ▸Companions: installable AI skill layers
- ▸End-to-end automated provisioning
- ▸Smart model switching across 600+ models
- ▸Entertainment Buddy companion in alpha
bskts
DeFi ExchangeCo-founded and built a live trading app for tokenized index baskets: real-time NAV across 15 baskets, candlestick charts, one-click gasless buys with per-leg fill feedback, keeper-filled limit, TWAP and bracket orders, a Weekend Cover market, and basket perps. Social login onto a smart account hides the wallet entirely, and a typed SDK and MCP server let AI agents trade on a user's behalf.
Highlights
- ▸Live on Robinhood Chain mainnet
- ▸15 baskets + the BSKTS index token
- ▸Gasless accounts, one-signature buys
- ▸Public SDK + MCP server on npm
onchain-ui
Open Source / Design SystemOpen-source shadcn registry: a component library and distribution model for teams building modern web interfaces, currently focused on web3 primitives (address display, token logos, prices, balances, network badges). Registry-first architecture, components install straight into your codebase via the shadcn CLI, inspectable and wired to your own data layer, not hidden behind a package boundary.
Highlights
- ▸Merged into the official shadcn registry directory
- ▸Live registry at onchain-ui.dev
- ▸One-line install via the shadcn CLI
- ▸Registry contract tests, docs with live demos
Web3 Product Suite + INDX
DeFi ProtocolCo-founded and led engineering on a full product suite, including swap, bridge, staking, yield farming, and an NFT platform, handling real-time data, wallet state, and transaction feedback under sub-second response requirements. Spun out INDX, a novel index protocol in the DeFi space, built on the same performance-first architecture.
Highlights
- ▸$1M+ revenue
- ▸4 chains deployed
- ▸~$50K TVL at peak
- ▸Community in the thousands
Algo Trading System
Systems / TradingAutonomous trading system for HyperLiquid perpetual futures, and the MVP prototype for a future ClawOps Trading Companion. Trade via natural language through Telegram, or let it run: every 4 hours it scores RSI, funding, and momentum signals alongside a DBRCI breakout scanner, then decides whether to act. Full execution layer with leverage, GTC limit orders, fill polling, and TP/SL bracket management.
Highlights
- ▸Natural language trading via Telegram
- ▸Autonomous 4-hour signal cycle
- ▸Full execution layer on HyperLiquid
- ▸ClawOps Trading Companion prototype
All built and designed by me.













