I'm Sam Sabey — 30 years building production software, now fully retooled for AI-native development. I build the systems that bridge the gap between what AI promises and what your business actually needs.
Based in Melbourne. Working with businesses across Australia and internationally.
Let's talk about what's possibleYou've seen what AI can supposedly do. Maybe you've tried to implement something and it didn't stick. Maybe you haven't started because you're too busy running your business to figure out where to begin. Either way, the gap between the hype and a working system that actually saves you time or money — that gap is where most businesses get stuck.
I've spent two years closing that gap. Not with demos or prototypes. With production systems that run, that scale, and that you can rely on.
I don't hand you a strategy deck and walk away. I build things.
Thirty years of shipping software — from C in the '90s through enterprise architecture across Asia Pacific to founding and building two startups from scratch — taught me what works in production and what doesn't survive first contact with real users.
When AI arrived, I didn't bolt it on. I went deep. Rebuilt how software gets designed, built, and delivered. Built my own orchestration tooling for managing AI-assisted development. Developed a structured, spec-driven process that produces consistent, quality results faster than traditional development.
The result: I build in weeks what used to take months. And everything I build is designed to run in production from day one.
Not AI as an afterthought. Architecture decisions made with intelligence built in.
Nothing ships that can't run reliably. No science projects. No eternal prototypes.
I personally do the work. You're hiring a senior engineer, not a firm that delegates to juniors.
Structured methodology that keeps scope tight and quality high, powered by the same AI tools I'm building for clients.
Full-stack production systems with AI capabilities built in from the ground up. Not a chatbot bolted onto your existing app — intelligent systems designed around your actual business logic. Rails, Python, modern web frameworks — whatever the problem demands.
Services as software. That thing your team spends hours coordinating manually every day? I build agentic systems that run it — monitoring, deciding, notifying, acting — with humans in the loop where judgment matters. Multi-platform integration, scheduled automation, intelligent alerting across the channels your team already uses.
Your documents, manuals, and knowledge base should answer questions — accurately, with sources you can verify. I build citation-fidelity retrieval systems where every answer traces back to the exact source. When the answer matters, "probably right" isn't good enough.
Eight years of connecting physical and digital systems — custom hardware, sensor networks, cloud platforms, device management at scale. If you have systems that don't talk to each other, APIs that need bridging, or IoT devices that need a brain, this is where decades of integration experience pays off.
You know AI matters. You don't know where it fits in your business. I come in, understand your operations, your data, your systems — and identify specifically where AI delivers measurable value. Then I build it. Not a report. Working software.
I practice what I deliver. These are live projects I'm actively building — they're how I stay sharp, prove out new approaches, and push the boundaries of what's possible with AI-native development.
Document Intelligence Platform
A citation-native retrieval platform where getting the source right isn't optional. Ask a question, get an answer, verify exactly where it came from. Built for professionals who need to trust the answers — not just get plausible ones. Production RAG with citation fidelity as the core architectural principle.
Agentic Orchestration
An agentic platform that monitors external services, detects changes, and routes intelligent notifications across multiple channels. Integrates browser automation, API discovery, scheduled scanning, and alerting — demonstrating the "services as software" approach in practice. A manual monitoring process, turned into an autonomous system.
AI Development Infrastructure
I'm not just using AI coding tools — I'm building the orchestration layer that manages them. Multi-session task management, context tracking across projects, structured workflows that maintain quality at speed. This is the fabrication workshop for making software in the modern age.
Sam Sabey
Founder, otageLabs
Melbourne, Australia
I wrote my first production C code in 1995, building energy trading systems on Sun Solaris. That was the era of dial-up modems and Oracle 7 — almost adjacent to the metal.
Through the late '90s and 2000s, I climbed the enterprise architecture ladder. Lead Solution Architect for a global software vendor, managing teams and bids across Hong Kong, Singapore, Taiwan, and Shanghai. A million frequent flyer miles in two years. Managed the architecture for a $95M billing system project. Presented at international conferences.
Then I realised I'd got so far from the technology that I'd lost the connection to building things. The titles were impressive. The distance from code was vast.
So I went back to building. Founded SmartEnergyGroups — an energy management startup where I built everything from open-source sensor hardware and firmware through to the web platform and analytics. Then eight years at ICU Solarcam, designing and building a national IoT video security platform from device to dashboard — custom gateways, thousands of managed devices, AI-powered alarm verification.
Two startups. Built from scratch. Hardware to software to operations.
When AI arrived, I didn't add it to the toolkit — I rebuilt the toolkit. Retooled how I design, build, and deliver software. Built my own orchestration systems. Failed at it, learned from it, iterated until it worked. That investment — the unglamorous, time-consuming, deep retooling that doesn't show up in YouTube tutorials — is what I bring to every engagement.
otageLabs is the embodiment of samotage — thirty years of building, breaking, learning, and building again. Now applied to the most significant shift in how software gets made since the internet itself.
46% of enterprise AI proofs of concept get scrapped before production. The PoC has become a performance that proves AI can do something without ever proving it will. Build the smallest real system instead.
OpenClaw dominates OpenRouter's Claude 3.5 Sonnet token chart, but total demand hasn't grown; it has been displaced. The early-adopter wave may already be cresting.
AI builds software faster than ever, but the bugs scale with the velocity. The gremlins are already on board — the question is what happens when they hit the water.
Whether you haven't started with AI yet or you've tried and hit the wall — I'd like to hear what you're working with. No pitch. No pressure. Just a conversation about what's possible for your business.
Based in Melbourne, working with businesses across Australia and internationally. I'll get back to you personally.