How to adopt AI in a small business in Singapore: the 5-step path we run
A practical adoption order for a small business — start with one low-risk process, keep the human checkpoint, expand only on evidence. The path we run ourselves.
Notes
What we've learned building and running software with AI — written from real projects, including the parts that went wrong.
AI readiness assessment for SMEs in Singapore: the 7 checks that actually matter
Skip the maturity score. The 7-point checklist we run before handing any business process to AI — and what each green light looks like in practice.
When AI-written code passes every test and still doesn't work
A feature shipped with a green test suite and did not work on real hardware for 19 days. What that taught us about which claims from an AI agent count as evidence, and which are just confident output.
AI vs offshore development cost — the maths changed
Offshore development was rational when senior local engineers were unaffordable. Here's what actually made up that cost, and what changed when AI took over the volume work.
Where AI gets it wrong — and how you find out
The failure mode isn't a wrong answer. It's a confident, well-formatted, entirely plausible answer that nobody thinks to check.
The quiet cost of not having a specialist
Small companies rarely fail from one big mistake. They lose ground slowly, in all the places they could never afford to hire.
Seven offshore developers to zero — and we got faster
In October 2024 we had seven offshore developers. Today we have none, and we ship faster. The dates, the numbers, and the one thing that makes it work.
What AI actually changes for a small business (and what it doesn't)
Most AI advice for SMEs is written for companies with a hundred staff. Here is the version for a company with fifteen.