Founding Engineer at Speakeasy, building the AI Control Plane that lets enterprises govern and secure their AI agents. Creator of AgentLink, an open source coding agent harness for VS Code.
I help teams adopt AI safely and productively, and give growing companies senior technical leadership without a full time hire. Based in Cairns, working with teams worldwide.
Principal Engineer building AI infrastructure, and a consultant helping teams put AI to work
Based in
Cairns, Australia
Where the 🌴 Rainforest meets the Reef 🐠
I'm a Principal Engineer and consultant based in Cairns, Australia. At Speakeasy I build the AI Control Plane that enterprises use to govern and secure their AI agents, and in my own time I build AgentLink, an open source harness for coding agents.
With 17+ years of engineering and technical leadership across developer tooling, FinTech, and online games, I help organisations move from AI experiments to AI that is part of how they work. That means enabling teams to use coding agents well, building agent systems that hold up in production, and providing fractional CTO leadership when a company needs senior technical direction without a full time hire.
17+
Years Experience
18
Largest Team Led
6
Teams Led
AI enablement, fractional technical leadership, and hands-on agent engineering
Adopt AI with confidence
Help your teams get real value from AI tools and coding agents, with the guardrails to do it safely.
Find where AI will actually move the needle, choose the right tools and models, and plan a rollout your teams will use.
Set up Claude Code, Codex, Cursor, and harnesses like AgentLink with shared AGENTS.md, skills, MCP servers, and review practices.
Guardrails for agent access, data exposure, prompt injection, and auditing, informed by building an enterprise AI Control Plane.
Hands-on workshops on AI directed development, so engineers ship faster without giving up quality.
Senior technical leadership, part time
Experienced technical leadership for startups and growing companies that need direction, not another full time executive.
Turn business goals into a technical plan, including where AI fits and where it does not.
System and code reviews, build versus buy decisions, and technical due diligence for investors and acquirers.
Hire your first engineers, set up processes that scale, and mentor the leads who will run the team.
A trusted technical voice for founders, from MVP planning through to scaling decisions.
Hands-on builds that ship
When you need AI systems built, not just advice. Production-grade agents, integrations, and platforms.
Tool-using agents and automated workflows designed to be reliable, observable, and cost aware.
Expose your APIs, data, and internal tools to AI agents safely through the Model Context Protocol.
Bring LLM features from prototype to production with the monitoring, security, and fallbacks they need.
Golang and TypeScript backends, OpenAPI-driven APIs, and cloud platforms on AWS and GCP.
A few hours a month of on-call advice, reviews, and decision support.
A focused engagement of a few weeks to roll out AI tooling and workflows across a team.
One or two days a week as your CTO or engineering lead, embedded with your team.
Scoped builds of agents, MCP servers, and AI features, delivered end to end.
Not sure which fits? Let's talk about where you are and where you want to be.
Worked with me before? I'd love to hear your feedback!
Write a Review on GoogleA coding agent harness built into VS Code
AgentLink runs frontier coding agents with the editor's intelligence, visible execution, reviewable changes, and as much or as little supervision as the task needs.
A capable model is only part of a capable coding agent. It also needs a good working environment: editor intelligence instead of text guesses, fast feedback instead of a surprise broken build, and a way for you to supervise or redirect work without taking the controls away.
Install
curl -fsSL https://raw.githubusercontent.com/reefbarman/agentlink/main/scripts/install.sh | bashAgents get language server navigation, diagnostics, symbols, and workspace-aware rename. Edits arrive as native diffs you can accept, reject, or adjust.
Review every action on risky work, or let familiar tasks move faster with focused rules and Approve for Me. Checkpoints make experiments reversible.
Tool calls, approvals, questions, and background agents stay visible in the chat and Activity Shelf, with a browser remote for supervision on the go.
Route review to a different model provider or external ACP agent, so important work gets an independent set of blind spots.
From game engines to AI infrastructure, building software at scale
The languages, platforms, and AI tooling behind my work
10+
Companies
5+
Industries
18+
Team Size Led
3
Cloud Platforms
Technical insights on software engineering, AI development, and building production systems
As coding agents get more capable, the craft shifts from writing every line to shaping the tools and feedback loops around them. Here's why I built AgentLink inside VS Code.
The speed of our OpenAPI linter wasn't achieved through profiling and micro-optimization. It came from designing the right abstractions from day one. Here's what I learned building a linter that processes 81K lines in 1.7 seconds.
OpenAPI is often dismissed as just documentation. But teams that treat it as infrastructure unlock automation, consistency, and scale across SDKs, docs, testing, and platform tooling.
Whether you want to roll out AI across your teams, need a fractional CTO, or have an agent system to build, I'd love to hear from you.
Based in Cairns, I work remotely with teams worldwide on AI enablement, fractional CTO engagements, and agent engineering. Let's discuss how I can help.
