Richard White recently told his staff at WiseTech Global that their craft is obsolete. It’s a bold claim from the billionaire founder of Australia’s biggest tech success story. He isn’t just talking about a minor shift in how we write code. He’s telling his developers that the very act of human programming is a relic of the past. If you’re a developer at WiseTech or any other firm pushing "AI-first" development, you’re likely wondering if your mortgage-paying skills just turned into a digital typewriter.
The reality isn’t as simple as a CEO’s memo. White’s stance reflects a broader, more aggressive push within the industry to slash overhead by automating logic. But there’s a massive gap between generating a snippet of Python and maintaining a global logistics platform that handles millions of transactions. Software engineering has never been about typing. It’s about solving problems.
The WiseTech AI Mandate and the Human Cost
WiseTech recently made headlines for its internal pivot toward total AI integration. Reports indicate that staff were told that AI is now "better than humans" at coding. This wasn’t a suggestion. It was a fundamental shift in the company’s philosophy. Employees described a culture where the traditional craft of software engineering was being devalued in favor of prompt engineering and automated oversight.
It's a high-stakes gamble. WiseTech builds CargoWise, a platform that manages global supply chains. This isn't a simple mobile game or a basic web app. It’s a massive, interconnected system where a single logic error can cause shipping delays across continents. Relying on Large Language Models (LLMs) to write the core of this system assumes that AI understands the nuance of international trade law, local customs regulations, and the messy reality of physical logistics. It doesn't. AI understands patterns in data.
When a company tells its workers their craft is dead, it creates a vacuum of accountability. If the AI writes the code and the human just "reviews" it, who owns the failure when a system crashes at 3:00 AM on a Sunday? The psychological impact on a workforce is immediate. You lose your best senior talent first—the people who actually enjoy the "craft" and don't want to spend their days babysitting a black box.
Why Coding and Engineering Are Not the Same Thing
The biggest mistake Richard White and other tech leaders make is conflating "writing code" with "engineering software."
AI is incredibly good at the former. If you need a function to sort a list or a script to scrape a website, ChatGPT or GitHub Copilot will do it in seconds. In that specific, narrow sense, the "craft" of memorizing syntax is indeed becoming less valuable. But that’s like saying the craft of a master architect is obsolete because we now have better power saws.
Software engineering involves:
- System Design: Understanding how a change in the billing module affects the warehouse tracking database.
- Security: Identifying subtle vulnerabilities that an AI might accidentally introduce because it saw similar (but insecure) code in its training data.
- Domain Expertise: Knowing why a logistics company in Singapore handles taxes differently than one in Rotterdam.
- Maintenance: Fixing a bug in a ten-year-old legacy system that doesn't have enough documentation for an AI to "understand" the context.
White's assertion that AI is better than humans ignores the "hallucination" problem. In a sandbox, a hallucination is a funny quirk. In a global logistics platform, a hallucination is a catastrophic financial liability. The idea that you can replace a seasoned engineer with a junior dev using an AI tool is a myth that will likely lead to massive technical debt.
The Productivity Trap of Automated Development
The push for AI at firms like WiseTech is driven by a desire for "hyper-productivity." The theory is that if an engineer can produce five times more code using AI, the company becomes five times more valuable.
It doesn't work that way.
More code usually means more bugs. It means more surface area to defend against hackers. It means more complexity for the next person to untangle. We’re already seeing "code bloat" become a major issue in organizations that leaned too hard into AI generation. The AI doesn't care about elegance or long-term maintainability. It cares about satisfying the prompt.
If you’re a WiseTech employee, you’re currently in a limbo state. You’re being asked to use tools that the leadership claims are better than you, yet you’re still the one who gets fired if the system breaks. This "human-in-the-loop" model is often just a way for companies to keep a legal scapegoat while trying to reap the rewards of automation.
The Risks of Devaluing Senior Talent
What happens when you tell your senior engineers they’re obsolete? They leave.
The tech industry is currently seeing a "hollowing out" of the middle. Companies are firing mid-level developers and expecting AI to fill the gap, while keeping a few "rockstar" seniors to oversee everything. But who trains the next generation of seniors? If junior devs only learn how to prompt an AI, they never develop the fundamental mental models required to debug complex systems when the AI fails.
WiseTech is risking its "institutional knowledge." When you’ve been building a product for twenty years, the value isn't in the lines of code. It’s in the brains of the people who know where the bodies are buried in the codebase. By telling these people their craft is dead, you’re basically inviting them to take that knowledge to a competitor who values human intuition.
How to Protect Your Career in an AI-First World
If you’re working in an environment that touts your obsolescence, you don't just sit there and wait for the axe. You have to pivot. Not away from code, but deeper into the parts of the job AI can’t touch yet.
Stop identifying as a "coder." Start identifying as a "systems architect" or a "product strategist." The value has shifted from how to build to what to build and why.
Focus on these areas:
- Requirement Refinement: AI is terrible at figuring out what a client actually needs when the client is being vague. Learning to translate messy human needs into rigid technical specifications is a superpower.
- Security Auditing: As AI-generated code floods the world, the demand for people who can find the "hidden" vulnerabilities in that code will skyrocket.
- Complex Integration: Making different platforms talk to each other is still a nightmare. AI struggles with the "edge cases" of API integrations and legacy hardware.
The "craft" isn't dead. It’s just changing. The people who will survive the WiseTech era of AI hype are those who realize that tools are just tools. A hammer didn't make the carpenter obsolete, and LLMs won't make the engineer obsolete—unless the engineer refuses to pick up the hammer.
Moving Forward in the AI Era
Don't buy into the "obsolete" narrative too quickly. CEOs have a vested interest in making you feel replaceable; it keeps wages down and stock prices up. While Richard White is busy telling the world that AI is the new master of the craft, the reality on the ground is much messier. Software is still a human endeavor because it serves human needs.
If you’re at a crossroads, start by diversifying your skill set. Experiment with the AI tools, but maintain your "low-level" knowledge. Understand how the memory is managed. Know how the network protocols function. When the AI generates a piece of code that looks correct but fails under load, you’ll be the only person in the room who knows why.
The best move right now is to become the person who manages the AI, rather than the person the AI replaces. That means leaning into the "human" side of engineering—communication, ethics, and long-term strategic thinking. If your company thinks you’re obsolete, prove them wrong by solving the problems their AI doesn't even know exist.
Update your portfolio. Document your wins that involved "un-breaking" an AI’s mistake. Stay sharp on the fundamentals. The hype cycle will eventually settle, and when it does, the companies that burned their human capital will be the ones left behind.
