The challenge
People researching a move to Canada face dozens of programmes with rules that change often. Existing tools either oversimplified the options into uselessness or buried people in the same jargon they were trying to escape.
Anything that guessed was worse than nothing, because people make irreversible decisions on this information. A confident wrong answer about someone's immigration status is not a bug, it is harm.
Our solution
We built the product end to end: a deterministic rules engine for eligibility, a guided assessment, and AI-written explanations grounded in official guidance with a source reference on every claim.
The interface is deliberately calm and shows its working: where you stand, which pathways fit, what to do next, and what it does not know. We also built the content architecture that makes it findable by search engines and quotable by assistants.
What we built
- Guided eligibility assessment with a deterministic rules engine
- AI explanations constrained to official sources, with references
- Personalised pathway reports
- A content hub engineered for organic acquisition and AI citation
- Structured data and a maintained llms.txt
The results
Eligibility is decided by a rules engine and only explained by a model, which is what makes an answer here safe to act on.