The SMH Business headline “Anthropic brings its mega IPO to Australia’s AI-starved investors” points to one side of the market, and the Computer Vision Engineer shortage sits on the other. I am keeping to what the headline says, because I have no inside detail on the IPO itself. The phrase “AI-starved” is what stuck with me. If investors here are starved for AI exposure, employers are starved for the people who build it, and computer vision is one of the thinnest corners. It also shapes the computer vision engineer talent shortage in Australia and the candidate expectations that now come with it.
What I am seeing at Big Wave Digital is a long, quiet search becoming normal for specialist engineering roles. There is no flood of applicants and no drama. A role opens, a handful of credible people are approached, most are already engaged on interesting work, and weeks go by between meaningful conversations. Companies that expect a four-week process get restless by month two and often quit by month four.
One rainy Sunday evening, Rach and I were watching the Winter Olympics from Milan. Rach used to work as a manager at Channel 7 on earlier Winter Games, so she tends to see the stories behind the broadcast. Eddie the Eagle and Steven Bradbury came up, two athletes remembered because they stayed in the race while others fell away. I thought about the 9-month Python/Django search we ran, and the 7-month paid media role Jules finally filled. Neither was a failure. Both were what staying in the race looks like when the talent is thin.
Why is an AI-starved investor market bad news for hiring?
When capital chases a theme, the first thing it does is raise the number of companies that want to build in that theme. Funding creates job specs faster than universities, bootcamps or career changers can create people to fill them. Investor appetite for AI is arriving in Australia faster than the supply of people who can build it, so every well-funded team ends up competing for the same small group of engineers.
Founders and CTOs tend to read a story like the Anthropic one as good news for the ecosystem, and in many ways it is. More attention on AI should mean more local investment and more ambitious products. My read is that the scarce resource has changed. A few years ago the question at board level was whether you could raise. Now the sharper question is whether you can staff the thing you raised for.
Computer vision feels this earlier than most disciplines. It sits between research and production engineering, and the people who are good at it have usually spent years on narrow, hard problems such as detection, segmentation, video pipelines and edge deployment. You cannot retrain a general software engineer into that in a quarter. The talent pool is small to begin with, and heavy AI investment pulls on it from several directions at once.
What is driving the Computer Vision Engineer shortage in Sydney?
The Computer Vision Engineer shortage in Sydney is the product of several things stacking up, and none of them clears quickly. The first is training lag. Strong computer vision engineers usually combine a research-heavy background with production experience, and that combination takes years to build. Demand can jump in a funding cycle, but supply moves on a much slower clock.
The second is competition from offshore employers. Many of the best engineers in this space can work for global teams remotely, and they know it. A Sydney company is not only competing with the firm down the road. It is competing with whoever has the most interesting dataset anywhere in the world. That changes how candidates weigh a local offer.
The third is how many industries now want vision capability. Retail, logistics, agriculture, mining, health, security and construction all have use cases, and most of them are hiring their first or second specialist rather than their tenth. That makes the search unusual, because the hiring manager often cannot lean on an established team to judge candidates. If you want a wider view of national labour conditions, the ABS labour statistics give the broad picture, though niche engineering disciplines rarely show up cleanly in them. You feel them in how long searches run.
This is where computer vision hiring in Sydney differs from, say, a general backend search. With a backend role, patience helps. With computer vision, patience is the strategy, and the people who tolerate a slow process well are the ones who end up with the hire.
4 things strong computer vision candidates now expect before they’ll talk to you
Candidate expectations have shifted, and I hear the same filters repeated across conversations. Strong engineers are not filtering on brand or perks first. They are filtering on whether the work is worth their time. These are the four I hear most.
- A genuinely hard, well-defined problem. Engineers want to know what you are trying to detect, measure or predict, and why it is difficult. A vague line about “using AI to improve the product” tells them nobody has done the thinking yet. A specific problem, even an unsolved one, signals a team that knows what it is doing.
- Real access to data. Models are only as good as what they are trained on. Candidates ask who owns the data, how it is labelled, how much exists and whether it is clean enough to build on. If the honest answer is “we are still collecting it”, say so early, because they will ask.
- Adequate compute. Engineers want to know they will not be waiting days to run an experiment. Training budgets and GPU access are now a standard part of the first conversation, and a team that cannot answer clearly loses credibility fast.
- A clear path to production. Many strong people have spent time on prototypes that never shipped, and they do not want to repeat it. They want to know how a model moves from notebook to deployed system, who owns that pipeline and whether the business will actually use what they build.
If your role cannot answer these four questions well, it is the role that needs work before the search does. This is the part of computer vision engineer candidate expectations that many hiring managers underestimate, because they assume salary and title carry the pitch. For this group, the quality of the problem carries it.
What computer vision hiring looks like when it goes well
The searches that land are rarely the fastest. They are the ones where the employer decided early what actually matters. When I sit down with a CTO or a founder at the start of a computer vision search, the most useful exercise is separating the must-haves from the nice-to-haves. Does the person need production deployment experience, or is strong research depth enough because you have platform engineers to carry it to production? Is a specific framework essential, or is the underlying skill what you are buying?
Teams that skip this step tend to keep moving the goalposts. They interview a strong candidate, decide the profile should be slightly different, and restart. Four months in, they have seen plenty of people and committed to none. In a thin market that is expensive, because the good candidates you passed on were not sitting idle for long.
The other common pattern in successful computer vision hiring is speed once the right person appears. A long search does not mean a slow decision. The companies that win have already agreed internally who needs to meet the candidate, what the process looks like and what they are prepared to offer. When the right engineer surfaces after months of quiet, they move quickly, and the gap between a hire and a miss is often a week.
Employers who treat a nine-month search as a failure tend to quit before the right person surfaces, which is the opposite of what the numbers suggest. The longer timeline is a feature of the market, not a verdict on your company. If you want a benchmark for how hiring demand is moving across the economy more broadly, SEEK’s employment insights are a reasonable place to start, though specialist roles like this one tend to run slower than any headline average.
Staying in the race: what a long search teaches about specialist hiring
Back to that Sunday evening. Steven Bradbury did not win his short track speed skating gold by being the fastest skater on the ice. He stayed upright and in the race while the leaders fell, and he was in position when it counted. Eddie the Eagle is remembered for a similar kind of persistence, showing up and finishing when the odds said he should not be there at all. Neither story is about talent in the usual sense. Both are about not dropping out.
That is what the 9-month Python/Django search and the 7-month paid media role Jules finally filled taught me. I will not dress up details I do not need to share, but the duration alone says a lot. For a long stretch of both searches, nothing visibly happened. There was no shortlist to celebrate and no offer to chase. The clients who stayed disciplined kept the role open, kept the brief honest and kept talking to the right people, and eventually the right person surfaced. The ones who lose these searches tend to be those who stopped at month four and restarted with a lower bar.
Staying in the race does not mean doing nothing for nine months. It means holding the standard while adjusting the things you can control. You can sharpen how you describe the problem. You can speed up your interview loop. You can be clearer about must-haves, and you can ask whether a passive candidate, someone not looking, would find your role interesting if a recruiter called them tomorrow. Those changes are in your hands even when the supply is not.
Frequently Asked Questions
How long does computer vision hiring usually take in Sydney?
For senior or specialised roles, plan on a longer search than your usual engineering hire. Some close in a few months, and others run well past six. A nine-month search is not unusual for a niche discipline, so budget your timeline and your internal expectations accordingly rather than treating month four as a warning sign.
Is the Computer Vision Engineer shortage likely to ease soon?
I do not expect it to ease quickly. Training pipelines are slow, and investor attention on AI keeps adding new employers to the pool of buyers. It may soften at the junior end as more graduates arrive, but experienced engineers who can take a model into production will stay scarce for some time.
What do computer vision engineers care about most when choosing a role?
The problem, the data, the compute and the path to production. Salary matters, but it usually follows the quality of the work in importance. Engineers want to see that the problem is worth solving and that the business can support the result being shipped.
Should we hire a generalist ML engineer instead?
Sometimes that works, particularly when the use case is well understood and you have strong platform support. If the project depends on deep expertise in areas like video, 3D or real-time inference, a generalist can leave you with a gap that shows up six months in. It helps to decide what depth you actually need before you start.
What this means for your next computer vision hire
The Anthropic IPO story is a reminder that money is moving toward AI faster than people are. If you need a computer vision hire this year, the practical response is to plan for the market as it is. Budget for a longer search, and decide your must-haves early so you are not rewriting the profile in month three. Put real effort into the problem you are offering before you chase the person, because the best engineers will filter on it before they filter on anything else.
One thing you can do this week is audit the role itself. Read your job description and your first outreach message as if you were a strong computer vision engineer who is not looking. Would the problem excite you? Would you know what data and compute you would have, and where the work ends up? If the honest answer is no, fix that first. Then stay in the race, because in a thin market, the ones still standing at month nine are often the ones who make the hire.
The future is bright, let’s go there together!
Thanks for reading,
Cheers Keiran
Big Wave Digital.
Born in Sydney. Built for digital.
Obsessed with tech.
Trusted by the best.
And, most importantly, ready when you are.
“Courage is knowing what not to fear.”
— Plato
Fear slow hires.
Fear bad hires.
Fear wasting time.
But don’t fear reaching out.
We’re right here.
Let us help you build a Brilliant team in Digital.
Big Wave Digital are experts in Digital Recruitment Sydney
At Big Wave Digital, Sydney’s leading digital, blockchain and technical recruitment agency, we have deep connections, experience and proven expertise, and the ability to achieve a win for all parties in the challenging recruiting process. We can connect to highly coveted digital and tech talent with the world’s best employers.
Keiran Hathorn is the CEO & Founder of Big Wave Digital. A Sydney based niche Digital, Blockchain & Technology recruitment company. Keiran leads a high performance, experienced recruitment team, assisting companies of all sizes secure the best talent.
Digital Marketing Recruitment in 2026 Sydney
