Your best people can't be everywhere. Their judgment can.

Every business runs on expertise that took years to build — the instincts, the judgment calls, the "we just know" that separates good from great. Most AI has no access to any of it. Orbital builds AI systems grounded in how your business actually works.

I'm Johan, an AI systems engineer in Christchurch, New Zealand. I help businesses take the operational intelligence they've accumulated — in their people, their processes, their hard-won decisions — and make it available at scale through AI that reflects their standards, not someone else's.

Why Most AI Projects Disappoint

Generic AI knows what's publicly true about your industry. It doesn't know what makes you good at what you do.

Your senior accountant's instinct for which deductions to flag. Your best lawyer's judgment about contract risk. The way your operations team triages exceptions before they become problems. None of that lives in a language model.

So the AI sounds plausible but flat. It gives textbook answers when your clients expect your answers. The automation handles the easy cases but misses the nuance that matters.

The fix isn't a better model. It's giving AI access to the expertise your business already has.

How I Work

I start from your business, not from technology. Every engagement follows three phases, and the first one stands on its own.

Map

Before building anything, I work with your team to surface the expertise that actually drives your business — the decision patterns, the judgment calls, the accumulated intelligence that lives in people's heads and informal processes but hasn't been made systematic.

We identify where high-value work is being done manually, where critical knowledge depends on specific individuals, and where AI would genuinely improve outcomes versus where it would just add complexity.

You walk away with: A clear map of your operational intelligence — where it lives, where the gaps are, and concrete next steps. This is valuable whether or not we continue working together.

Architect

With your expertise mapped, I design the systems that make it usable. This means structuring your operational intelligence so AI can draw on it, then designing how that connects to your existing tools and workflows.

The goal is AI output that reflects your team's standards and judgment — not a generic assistant wearing your logo. What gets structured, how it connects, what context the AI needs: all of this is specific to your business.

You walk away with: A system design — the intelligence architecture and the AI systems shaped around it. Could be custom-built systems, configured existing tools, or a mix.

Build

I build systems designed to last beyond the engagement. Your team should be able to maintain and extend the intelligence base themselves, so it gets more valuable over time rather than degrading.

You walk away with: Working systems your team owns and can grow — not a deliverable that starts decaying the moment I step away.

What This Looks Like In Practice

The best way to understand the approach is through what changes for the business.

An accounting firm whose senior partners' tax planning instincts now inform every client engagement — not just the ones they personally handle. Junior staff produce work that reflects the firm's accumulated judgment, not just technical compliance.

A law practice where decades of contract negotiation patterns are available to every associate. The AI doesn't practise law — it surfaces the firm's own precedents, risk patterns, and drafting standards when they're needed.

An environmental consultancy whose field assessment expertise runs through every site evaluation, so reports reflect the consultancy's specific methodology and regional knowledge rather than generic frameworks.

A logistics company where the operations team's exception-handling judgment is systematised. The AI triages disruptions the way your best dispatcher would, escalating the right things and resolving the routine ones.

These aren't chatbots. They're systems where the AI draws on what your business actually knows.

Who This Works For

Good fit: You have deep domain expertise and want to make it systematic. Your best people's judgment isn't scaling with the business. You're spending senior time on work that should be informed by your accumulated intelligence. You may not have a technical plan yet — that's what the Map phase is for.

Sectors where this lands hardest: Professional services (accounting, legal, consulting), environmental and engineering, industrial operations, financial services and compliance, healthcare, logistics and supply chain — anywhere institutional expertise is deep, hard-won, and trapped in individuals.

Not a fit: You want generic chatbots or AI wrappers. Off-the-shelf SaaS would solve the problem. You need hourly contractors or staff augmentation. If existing tools would work, I'll tell you that upfront.

What's Underneath

I'm a solo practitioner, not an agency — you work with me directly. My background is over a decade building production systems across fraud detection, industrial monitoring, scientific computing, and geospatial intelligence.

I build with React, TypeScript, Next.js, and modern AI infrastructure (LLM orchestration, MCP servers, evaluation frameworks). But the technology follows the problem. If the right answer is configuring existing tools rather than building custom systems, I'll say so.

Working style: Remote-first from Christchurch, New Zealand (APAC timezone, good overlap with Australia and Asia, async-friendly for US and Europe). Direct communication, fixed-fee engagements, outcomes over hours.

Start With A Map

The first step is a Map engagement — a structured process where we surface your operational intelligence, identify where AI has genuine leverage, and produce a concrete plan.

You leave with clarity on where your expertise is trapped, where AI will actually improve outcomes, and what to do about it — whether or not we keep working together.

If you're curious whether this fits your situation, email me. I respond to every serious inquiry within 24 hours.

Email: info@orbital.co.nz

Recent Writing

Why Your AI Chatbot Failed Most AI projects fail because they skip the context that makes them useful.

Building AI Products That Ship From insight to production in weeks, not months.

Building with MCP Connecting AI assistants to real infrastructure using Model Context Protocol servers.

Common Questions

How is this different from hiring an AI consultancy? Most consultancies lead with the technology and look for places to bolt it on. I start with your operational intelligence and figure out how to make it available at scale. The output isn't a strategy deck — it's working systems your team owns.

Can't we just use ChatGPT/Claude/existing tools? Maybe. If off-the-shelf tools solve your problem, I'll tell you that upfront. I work on problems where generic AI falls short — usually because the system needs to reflect your specific expertise, standards, and decision-making rather than general-purpose responses.

What does a Map engagement cost? It's a paid engagement, not a free consultation. Pricing depends on complexity and scope. Email me and I'll give you a straight answer.

Do you only do strategy, or do you build too? Both. The Map phase stands on its own, but I also architect and build the systems. The difference is that I won't build until we've properly understood what intelligence the AI needs access to and where it creates real value.

What if we already know what we want built? Great — we move faster. But I'll still want to understand the expertise driving the system, even briefly. It's the difference between AI that reflects your business and AI that could belong to any business in your sector.

Do you work with startups? Sometimes, if you're post-product-market-fit with genuine domain expertise to systematise. Early-stage companies still working out their own processes usually aren't ready for this yet.

Bottom Line

Your competitors can license the same AI models you can. They can't replicate your expertise.

The businesses that pull ahead with AI will be the ones that make their own judgment, processes, and hard-won intelligence available at scale — building something that gets more valuable over time, not more generic.

That's what Orbital builds.