Coding bootcamps in Australia: what the job outcomes actually look like

A coding bootcamp homepage is a study in confidence. A placement rate with a percent sign. A graduate in a hoodie on a beanbag. A salary that would rent a one-bedroom in Parramatta without flatmates. The junior market in 2025 and 2026 is a different document. It is full of people who finished last year, people who finished a computer science degree, and job ads that want two years of production React and the willingness to be paid like a first-year.

This is not a consumer warning dressed as a hatchet job, and it is not a defence of universities. It is a way to read outcome claims the way a sceptical careers counsellor — or a lawyer who has skimmed the Australian Consumer Law — would read them. The careers hub is the rest of the shelf. If the sales call has a countdown, read how to spot an online course scam before you give them a date of birth.

What a bootcamp is, in Australia, in practice

A typical full-time web or software bootcamp here runs something like 12 to 24 weeks, charges somewhere between $10,000 and $25,000 if you pay upfront, and promises a portfolio, a bit of career coaching, and a stack that was fashionable when the syllabus was written — often JavaScript, a modern front-end framework, a bit of Node, a cloud free tier, and a group capstone. Part-time versions take most of a year and cost about the same. Data and UX variants exist; they should be judged on the same outcome maths.

Some of these schools are registered training organisations and can issue an AQF qualification. Most are not. That distinction matters for funding, for visas, and for any employer who still thinks in diplomas. It does not tell you whether the teaching is good.

Campus postcodes have thinned since the last boom. A few brands still run rooms in Sydney and Melbourne. Plenty are online with “live” cohorts on Australian time, which is a real advantage over a US evening lecture, and a real disadvantage if you needed someone to look at your screen.

A bootcamp can work. People do get hired from them — into junior developer roles, into support-plus-scripting, into test, into analytics-adjacent work, into the same company that already employed them. The question is not whether a success story exists. The question is what the distribution looks like, and whether the school will show it to you.

The junior market in 2025–26 is not 2021

The last time bootcamp marketing copy was written in a hurry, Australian tech firms were hiring anyone who could spell “component” and remote US salaries were leaking into Slack screenshots. That market broke. What you have now, if you read SEEK and LinkedIn in Sydney, Melbourne, Brisbane and the smaller scenes in Adelaide, Perth and Canberra:

• Fewer graduate and junior seats than applicants. Offshoring and hiring freezes did not wait for your capstone.
• Ads that say “junior” and then list a year or two of production experience, AWS, and a take-home that would take a mid a full day.
• Internal mobility. A lot of “first developer jobs” are being filled by people who already sit in the building — QA, support, analysts who can script.
• AI tools that make a weak junior less necessary for the chores a junior used to be hired to do, without making a strong junior any less valuable. The floor rose. The brochure did not.

None of this means “never enrol”. It means a placement rate calculated on the class of 2021 is a historical artefact. Ask for the last two graduating cohorts, in Australia, after the market cooled. If the school cannot or will not separate those years, they are averaging a party with a hangover.

Salaries followed. A first developer seat in 2026 is often advertised from the low-to-mid $70,000s into the $90,000s in the big cities, more if you already have a domain (finance, health, government) or a degree. The $110,000 “junior” is not the median and it is not a promise. Contract work exists and will drop you on a Friday.

If you are changing from hospitality, retail or a non-technical degree with no adjacent desk, budget 6 to 18 months after the bootcamp, not “employed by graduation Friday”. People who land in week thirteen usually had a job in the next room, a portfolio before day one, or a definition of “in field” that includes unpaid internships and casual tech support.

Advertised placement rates, taken apart

A percentage on a homepage is a claim about trade. In Australia, claims of that kind sit under the Australian Consumer Law — misleading or deceptive conduct, and false or misleading representations about employment outcomes. You do not need to be a regulator to ask the questions the ACCC would find ordinary.

When a school says “87% of our graduates are hired”, ask:

1. Hired to do what? A junior software engineer role counts. A casual service-desk role at a retailer, a return to the student’s previous admin job, or a three-week internship the school arranged with a partner, should be labelled. “In field” is a phrase with a rubber boundary.
2. Hired by when? 90 days, 180 days, 12 months? A clock that starts after an “optional” career week can hide a delay.
3. Who is in the denominator? Did they exclude people who took a break, people who failed a final, people who did not complete an unpaid project, people who stopped answering emails? A 90% rate on a cleaned cohort can be a 55% rate on everyone who paid.
4. Where? Australian jobs, or global remote? A graduate in Logan who cannot move is not helped by a hire in Austin.
5. Self-report. Is this a LinkedIn scrape the school interprets generously, or a statutory declaration? People inflate titles. Schools do not always push back.

Ask for the table, not the round number. A school that is proud of its outcomes will show you cohort size, completion, “in field” definition, median salary, and the year. A school that is proud of its ads will show you a testimonial and a beanbag.

If they will only disclose under an NDA, you are not a student. You are a liability they have met before.

The desk’s piece on Australian Consumer Law for people who sell courses is written for providers. Read it anyway. It will show you what a careful seller is afraid to put on a landing page. That fear is information.

Income-share agreements are not a scholarship

An income-share agreement (ISA) says: we will teach you now; you will pay us a percentage of your salary later, for a fixed term, once you earn above a threshold, up to a cap. It is marketed as aligned incentives. It is also a credit-like product sold by a school that grades you.

Read the contract as if a friend were about to sign it after a long call.

• Threshold. If repayments start at $60,000, a $68,000 job in Adelaide has you paying from month one. A threshold that sounded kind in a Sydney salary screenshot is less kind in Townsville.
• Percentage and term. Four years at eight per cent of gross is a lot of money on a $85,000 salary. Do the multiplication. Compare it to a bank loan, a FEE-HELP place on a real degree, or paying the $15,000 now.
• Cap. Is the cap 1.5 times tuition, 2 times, or something that only looks like a cap? What happens if you take a job that includes super, bonus, or a contractor day rate?
• What counts as a job. Some contracts want their cut from any employment. Some only from “tech”. The definition will be argued when you are tired.
• Deferral and hardship. Get it in writing. Verbal kindness from an admissions person is not a clause.
• What happens if you quit, fail, or the school folds. Australian students have already lived through campuses that quietly closed a product line. An ISA with a defunct counterparty is a mess you do not want.

An ISA can be rational if you have no savings, you have a high chance of a qualifying job, and the cap is sane. It is a poor way to buy optimism. If the school needs the ISA to make the price disappear in the sales call, price the disappearance.

Pay-after-placement deals that are not formal ISAs can be worse: vague, undocumented, and suddenly due. If it is not a written contract you can take to a community legal centre, it is not a deal.

When a computer science degree still wins

A three-year CS or software degree (AQF level 7) is slower and, if you are paying full freight, can be dearer than a bootcamp. It still wins in rooms a bootcamp does not enter cleanly:

• Graduate programs at banks, insurers, Telstra-scale firms, the bigger consultancies, and much of government. The filter is often a degree. A portfolio helps. It does not always replace the filter.
• Roles that want theory you will not get in twelve weeks: compilers, operating systems, algorithms as more than a leetcode evening, maths for certain data and graphics seats.
• Research, machine learning beyond a tutorial, and anything adjacent to a university hospital or a defence science shop.
• Migration and licensing conversations that still score AQF levels. A bootcamp certificate is usually not a degree, even when the teaching is better than a particular university unit. Do not let a marketer blur that.
• A second chance at being “junior” when you are 22 and can afford the years. Age is not a moral category, but graduate programs have habits.

The degree does not automatically win against a bootcamp graduate who has shipped, who already works in the building, or who is applying to a twenty-person shop that wants a repo and a person who can sit with a customer. Plenty of CS graduates cannot ship. Plenty of bootcamp graduates can. The market is not a morality play. It is a set of filters, and some of those filters are lazy.

If you already have a degree in something else, a bootcamp or a targeted online path (see the desk’s piece on changing into data analytics for the analyst version of this argument) is often more rational than a second bachelor. If you have no degree and you can get a Commonwealth-supported place, run the numbers against a $20,000 bootcamp that does not come with a HELP loan or a student income bank.

Honours and master’s pathways are a different product again. Do not buy a master’s in IT as a disguised bootcamp unless you have checked the unit list and the graduate-program eligibility. Some are excellent. Some are a full-fee warehouse.

What the work sample has to be

Hiring managers who will look at a bootcamp graduate are hiring the repo, not the lanyard.

A group capstone about a fictional plant shop is what everyone else has. It signals that you finished the course. It does not signal that you can own a ticket.

Better evidence:

• A project you use. A small tool for a real workplace, a volunteer org, or your own ugly problem.
• Tests, a README that tells the truth, and a deployment a stranger can click.
• A pull request history that is not four giant commits on the last night.
• One piece you can explain at the level of trade-offs: why this state library, why this query, what you would rewrite.

Assessments that are multiple-choice or closed-repo tutorials will not travel. If the school will not let you publish, you do not own the work. Ask before you pay.

Career coaching is only as good as the person who has placed juniors this year in your city. A “lifetime Slack” full of people asking whether anyone is hiring is not a labour-market strategy.

How to interrogate a school in one week

You do not need a mystery shopper kit. You need a weekday and a spine.

• Sit in on a class, or watch an unedited hour. Sales webinars do not count.
• Speak to two graduates from the last 12 months who were not supplied on a sheet. LinkedIn exists. Ask what their title is and when they were hired. Ask what they would not do again.
• Get the outcome table in writing, with definitions.
• Get the refund, deferral and complaint path in writing. ACL rights sit underneath a contract; they are not replaced by it. A “no refunds, ever” clause is a review of their confidence.
• Price the stack you must own: a laptop that can run the tools, cloud credits, exam vouchers if they have bolted CompTIA or AWS onto the brochure.
• Ask who taught last month, and whether they are still in industry. A founder who last shipped in 2018 will teach you a stack that interviews badly.

If any of those requests are treated as hostility, believe them. You are about to hand over five figures.

Adjacent outcomes that are not failure

Not getting a “software engineer” title in six months is not the only honest result.

People leave bootcamps into test, support, implementation, junior data, Salesforce-adjacent work, and internal tools roles. Those jobs pay, they sit next to the code, and they are how a lot of Australians eventually become developers. A school that treats anything except “engineer at a scale-up” as failure is selling a fantasy about status.

People also leave into nothing, or into debt plus the old job. That is the tail the homepage crops. Budget as if you might be in that tail. If that budget is impossible, do not enrol. The school will not make rent while you job-hunt.

A working decision

Enrol if all of these are true:

• You can already learn from documentation for a week without a teacher, and you have a small project to prove it.
• You have 12 to 24 weeks (or a real part-time calendar) that will survive paid work and care.
• You have priced the junior market in your city in the last month, not in a testimonial.
• You can survive 12 months after the course without the advertised salary.
• The contract — upfront or ISA — is one you could explain to a parent who is bad at being impressed.

Do not enrol if the deciding fact was a percent sign, a fear of being “left behind”, or a visa story the school is not licensed to give you.

A bootcamp is a compressed, expensive, sometimes excellent way to practise. It is not an apprenticeship, not a degree, and not a job. The outcome is the work you can show and the market you walk into. Everything else is advertising. Read it like advertising. Then, if you still want the labour, buy the labour — not the beanbag.