The narrative says there’s a massive AI talent shortage in Australia. That’s partly true. But the full picture is more nuanced than “everyone is hiring and nobody can find anyone.” Let me break down what’s actually happening in the Australian AI job market.

What the Numbers Show

Job listings containing “artificial intelligence” or “machine learning” on Australian job boards grew by 67% year-over-year. That’s significant. But context matters: many of those listings are for roles that use AI tools rather than build them. A marketing manager who needs to use AI content tools isn’t an “AI job” in any meaningful sense, but it shows up in the statistics.

Genuine AI engineering and data science roles grew by about 35%. Still healthy, but less dramatic than headlines suggest.

The median salary for a mid-level machine learning engineer in Sydney is now around $170,000 to $190,000. Senior roles at major tech companies break $250,000 easily. In Melbourne, salaries run about 10-15% lower. Brisbane and Perth are catching up but remain 15-20% below Sydney.

What Employers Actually Want

Here’s where the disconnect lives. Many job listings ask for a fantasy candidate: PhD in machine learning, five years of production experience, expertise in PyTorch and TensorFlow and JAX, experience with LLMs and computer vision and time series, plus domain knowledge in [insert specific industry].

That person doesn’t exist. Or if they do, they’re at Google DeepMind earning twice what your company pays.

The companies that are actually hiring successfully have figured out what they genuinely need versus what would be nice to have. The most common realistic requirements for mid-level AI roles in Australia are: strong Python skills, experience with one major ML framework, ability to work with messy real-world data, and enough communication skills to explain technical concepts to non-technical stakeholders.

Domain expertise is increasingly valued over theoretical knowledge. A data scientist who understands Australian financial services regulation is more valuable to a bank than a brilliant researcher who’s never worked with business constraints.

Where the Jobs Actually Are

Financial services remains the largest employer of AI talent in Australia. The big four banks, major insurers, and super funds are all building internal AI capabilities. These roles tend to be well-paid and stable but can involve navigating significant bureaucracy.

Healthcare AI hiring is growing rapidly but from a smaller base. Harrison.ai, Appen’s health division, and several hospital networks are actively recruiting. The challenge is finding people who combine technical skills with an understanding of clinical workflows and health regulation.

Government is a surprisingly active AI employer. The ATO, Services Australia, Defence, and various state government agencies are hiring AI specialists. Government salaries are typically below private sector equivalents, but job security, work-life balance, and interesting problem spaces attract people who’ve burned out on startup pace.

Mining and resources companies are building AI teams focused on operational optimisation, predictive maintenance, and geological analysis. These roles often require travel to remote sites, which limits the candidate pool but increases compensation.

Startups offer equity and excitement but lower base salaries. The AI startup scene in Sydney and Melbourne is vibrant, but many early-stage companies struggle to compete on salary with established players.

The Skills Gap Nobody Talks About

The most acute shortage isn’t in AI researchers or engineers. It’s in AI translators: people who can bridge the gap between what AI systems can do and what businesses need them to do. Product managers who understand ML capabilities. Project managers who can run AI implementations. Business analysts who can identify automation opportunities.

These roles don’t require deep technical skills, but they do require enough technical literacy to have productive conversations with engineering teams and enough business acumen to quantify value. Australia has very few people with this combination, and formal training pathways for it barely exist.

What I’d Tell Someone Starting Out

If you want to work in AI in Australia, my honest advice is this.

Don’t chase another degree. A Master’s in AI can be valuable, but employers increasingly care about demonstrated skills over credentials. Build a portfolio of projects using real datasets. Contribute to open-source. Do freelance work if you can’t get a full-time role immediately.

Learn the boring parts. Data cleaning, pipeline engineering, model monitoring, and documentation. These aren’t glamorous skills, but they’re what separates someone who can build a Jupyter notebook demo from someone who can ship production AI systems.

Pick a domain. The generalist AI engineer market is competitive. The AI engineer who understands mining operations, or healthcare compliance, or agricultural supply chains, is rare and highly sought after.

And be realistic about timelines. Getting your first AI role in Australia typically takes longer than in the US because the market is smaller. But once you’re in, career progression is rapid because demand exceeds supply at every level.

The AI job market in Australia is genuine and growing. It’s just not quite the gold rush that some recruiters would have you believe.