Blog

For more than 18 years, Wei Khjan Chan has worked as an accountant, a profession often flagged as being at risk of automation. Each time he saw headlines warning that AI could replace jobs like his, he said he felt the pressure mounting.

“It’ll be great if I get to know AI earlier. At least I replace myself rather than let other people replace me,” the 39-year-old told Business Insider.

To stay ahead of the curve, Chan picked up vibe coding, using AI tools to write code and build apps. The audit partner at an accounting and advisory firm in Malaysia said he stumbled on vibe coding in June after attending weekend coding workshops in Singapore and Malaysia.

Despite having no technical background, Chan built a web app to solve a pain point in his professional life: filing expense claims after business trips.

The app uses AI-powered optical character recognition to scan and process receipts, automatically exporting them into files for his company’s finance teams. He’s also using AI to automate his workflow, such as generating invoices.

“This code is a bunch of JavaScript, which obviously I don’t understand,” he said, showing Business Insider his web app. “Without the vibe coding tools and the skill set, an accountant is unable to do this,” he added.

AI isn’t a ticket out of accounting — it’s how to save it

The accountant said he didn’t learn vibe coding to make a career switch. Instead, he sees “AI know-how” as a fundamental skill for any office profession, like Excel.

Building his own apps showed him just how powerful the tools can be: what once required weeks and an entire team to build as a proof of concept can now be prototyped in a single weekend, he said.

Chan also told Business Insider he’s advocating for broader adoption. As a committee member in his local accounting institute in Malaysia, he’s lobbying for more AI training at scale.

Fewer people are pursuing accountancy even as demand for accounting services rises. With manpower falling short, AI could help fill the gap, Chan said.

Lessons learned from vibe coding

Chan said that when he first started experimenting with AI, he was advised to write long, detailed prompts with “full context length.” But experience taught him that smaller, iterative steps work better.

Related stories

Business Insider tells the innovative stories you want to know

Business Insider tells the innovative stories you want to know

“The initial prompt is very important to set everything right,” he said. After that, when changes are needed, it’s more effective to adjust one small part at a time instead of piling on an entire wish list.

He approaches it like managing an intern: Break tasks into smaller, precise instructions. The more specific you are, the better the outcome, he said.

Not every lesson came easily. In one project, Chan built his database based on a single organization. When someone later asked for multi-company support, he realized he had to rebuild the entire structure.

“It’s a very fundamental change,” he said. “I messed up everything.”

The experience taught him that getting the architecture right at the start is critical because features and functions can always be layered on later.

As for debugging, it’s basically like “complaining to the AI,” Chan said with a laugh. If the error message changes, that’s usually a good sign — the AI is working through the problem. If the same error keeps coming back, he said he’ll reset the conversation and reframe the request with new examples.

And he said that, despite occasional debugging, vibe coding doesn’t require endless hours of grinding.

Chan usually tinkers after his kids go to bed, adding a feature here or refining a function there. “It’s like playing a game,” he said.

Over time, it builds up, and with a little guidance, the pieces eventually come together.

Do you have a story to share about vibe coding? Contact this reporter at cmlee@businessinsider.com.