For thirty years there was a gatekeeper at the door of every technology career: years of programming, a computer science degree, or both. Quietly, over the last three years, the gatekeeper left.
This is not a motivational claim. It is a description of how software is now made. AI writes most working code today, and the people building successful small applications are increasingly not classical programmers but people who can direct AI well: decide what to build, judge what the machine produces, catch its mistakes, and ship something real people use. Those are thinking skills, not typing skills, and they can be learned in months by adults who have spent their working lives doing something else entirely.
Why the timing matters, in four numbers
Career changes are risky when a field is crowded, and less risky when demand is racing ahead of supply. Consider where Canada sits in mid 2026. About 19% of Canadian businesses use AI, according to Statistics Canada, which means four in five have not started. In rural regions, including most of the Fraser Valley, adoption is under 10%. Ottawa's AI for All strategy is investing $2.3 billion to change that, with a stated ambition of majority adoption within a decade and hundreds of thousands of AI related jobs. Every business that makes that jump needs at least one person who understands both the tools and the business itself.
Nobody is training those people at anything close to the required rate, and almost nobody is training them outside the big cities. That gap between what the economy needs and who exists to supply it is what a career opportunity looks like from the outside.
What you actually need (and what you do not)
You do not need advanced math. You do not need to write code from scratch; you need to read and shape it the way a good editor works with a writer, which is a far smaller skill and learnable in weeks. You do not need youth: the scarce commodity in AI work is judgment, and judgment is what your years in retail, trades, healthcare, administration, or agriculture were quietly building. A former store manager understands inventory pain better than any computer science graduate, and when AI handles the code, understanding the pain is the differentiator.
What you do need is honest: ten to thirteen hours a week for a season, curiosity that survives frustration, and a willingness to ship imperfect things and improve them in public. If those three are true of you, the door is open.
The three realistic paths
- Freelance builder. Small businesses everywhere need websites, booking systems, quote calculators, and simple applications. With AI, one trained person delivers what once took a small agency. This path starts earning fastest and scales with reputation.
- AI consultant to small businesses. The four in five businesses that have not adopted AI will not be served by Toronto consultancies. They will be served by someone local who can look at their operations, find the friction, and implement the fix. This path pays for business judgment, and it is where career changers with deep work experience shine brightest.
- Junior developer, the AI era version. Employment in a team, building with AI daily. The title is evolving faster than job boards can rename it, and portfolios now beat credentials in most small company hiring.
On money, honesty first: no reputable school promises income. For context only, Canadian salary data puts AI consultants around $90,000 a year on average; that is a market statistic, not a guarantee, and individual outcomes vary with effort, market, and path.
The two ways to learn
The self taught route is real and free: online tutorials, documentation, and thousands of hours of YouTube. Its price is paid in time and attrition instead of dollars. Expect six to twelve months to a credible portfolio, and be honest with yourself about the completion statistics of self paced learning; the internet is littered with abandoned folders named "learn_coding". It suits people with strong existing structure and no deadline.
The structured route trades money for compression and completion: a cohort, a schedule, an instructor who has seen your exact confusion before, and the social gravity of classmates expecting you to show up. Twelve focused weeks with live teaching reliably outperforms a year of solitary evenings for most working adults, not because the content differs but because finishing does.
That second route is what we built the AI Builders Program to be: 12 weeks, live online across BC and Canada, evenings on Pacific time, no technical background required. Students design, build, and launch two real applications, learn maintenance, marketing, and how to get paid, and finish with a portfolio that exists on the actual internet. The Fall founding cohort begins September 15, tuition is $3,995 with every tool included, and applications close September 12.
The best first step costs nothing: we hold free online info evenings on August 20 and September 2 at 7pm Pacific, where the whole program, the fees, and the honest expectations are laid out live, and every question gets a straight answer. Save a spot here.
Common questions
Am I too old to start at 40 or 50?
No, and in this field the question is nearly backwards. The scarce skill is judgment about what to build and whether it is right, and judgment is what twenty years in any workplace builds. Your domain experience is an asset pure technologists cannot buy.
Do I need math or coding?
No advanced math. Coding at a reading level rather than a writing level: AI writes, you direct and judge. That reading level is learnable in weeks and is exactly how our weeks two and three are designed.
How long before it pays?
Honestly, it varies, and anyone promising a number by a date is selling something. A focused learner can be producing portfolio work in about three months and pursuing first paid work after that.
Can I do this while working full time?
Yes, if the format fits working life: part time, evenings, roughly 9 to 13 hours a week treated as fixed appointments. That constraint is exactly why our program runs the way it does.
Where to go from here
If the self taught road fits you, start tonight; the only cost of trying is an evening. If the structured road fits better, read the full program, and come to an info evening before deciding anything; a $4,000 decision deserves an hour of looking the school in the eye first. Either way, the gatekeeper is gone. The only person left at the door is you.
Sources: Statistics Canada, AI use by businesses, Q2 2026 · Canada's AI strategy (AI for All)