I'm a product engineer who builds things that work — from full-stack web apps to multi-agent AI systems that run autonomously. I care about shipping products that solve real problems, not just impressive demos.
Agentic AI
I design and operate multi-agent systems where autonomous AI agents coordinate, delegate tasks, and make decisions with minimal human oversight. My current stack includes orchestration layers that manage specialised agents — each with distinct roles like security auditing, trading execution, research, and code review — all running on scheduled workflows with automated monitoring and self-healing capabilities.
This isn't theoretical. I build systems that handle real workloads: automated grid trading with exchange API integrations, memory systems with integrity verification and provenance tracking, PII scrubbing pipelines, backup orchestration with disaster recovery testing, and security scanning that runs continuously in the background. The agents coordinate through structured protocols, escalate when confidence is low, and produce auditable outputs.
I'm particularly interested in agent reliability — making sure autonomous systems fail gracefully, maintain state correctly, and don't drift from reality. Financial APIs that break on float serialisation, cron jobs that silently stop executing, cached state that diverges from source of truth — I've debugged all of these and built systematic safeguards against each one.
Product Engineering
I'm based in Sydney, Australia, currently building digital products in the disability services industry — work that has a direct impact on people's lives. Before that, I worked at one of Indonesia's largest enterprise SaaS startups, and at an Australian travel media company building content platforms.
I'm not tied to any single stack. Right now I work mostly with Vue, TypeScript, Kotlin Spring Boot, and MongoDB — but I've shipped production code across JAMstack, headless WordPress, and cloud infrastructure tooling. I pick the right tool for the problem, not the trendiest one.
I care about infrastructure and security — not as afterthoughts, but as foundations. I believe in building systems that fail gracefully and maintain themselves reliably.
Beyond Code
Martial arts (武道) got me into Japanese and Korean culture, and I never came back. I speak conversational Japanese (日本語), have a working understanding of Korean (한국어) and basic Mandarin (中文). Learning languages rewires how you think about communication — which turns out to be useful when designing interfaces for both humans and AI agents.
I read widely — the history of democracy, East Asian civilisation, personal finance, and cybersecurity research. I find that the best engineering decisions come from understanding systems beyond just the technical layer.
This Site
I write to organise my thinking and share what I've learned. You'll find posts on personal finance, technical deep-dives, and the occasional creative piece. Everything here comes from personal experience — no sponsored content, no affiliate hustle.