Sertus — AI for Canadian labour relations.
A retrieval-grounded AI platform that searches collective agreements, flags relevant clauses, and tracks every grievance from filing through arbitration. In production with Canadian unions and HR teams.
The problem
Union stewards and HR teams spend hours digging through collective agreements, tracking grievance deadlines, and preparing for meetings. The work is detail-heavy and deadline-driven: miss a notice requirement by a day and the case is gone.
Most of that knowledge lives in people’s heads. When senior stewards leave or retire, institutional memory walks out the door with them. New staff start from scratch, and the same questions get re-researched every month.
What we built
Sertus is a retrieval-augmented AI platform built around the applicable collective agreement and the organization’s own case history. It searches the text, flags relevant clauses, generates meeting talking points, and tracks every grievance from filing through arbitration.
Every answer cites its source — the exact article and clause in the CA — so stewards can verify before they act. The system is designed to be auditable by people who are not technical, because that’s who uses it.
- Searches collective agreements and cites the exact clause
- Tracks grievances across filing, steward review, arbitration, and close-out
- Captures institutional case-handling memory so it doesn’t walk out the door
- Data stored in Canada, subject to Canadian privacy law
Outcome
Research that used to take hours now takes minutes. Stewards walk into meetings with talking points grounded in the actual agreement, not half-remembered precedent. New staff onboard faster because the knowledge lives in the system, not in a retiring colleague’s head.
Sertus is live and running in production.
Cites its sources
Every answer is grounded in the applicable collective agreement — article and clause.
Institutional memory
Past case handling is captured, searchable, and never lost to turnover.
Canadian data, Canadian law
Federal and provincial frameworks supported. Data stored in Canada.
Working on something like this?
We'll walk you through how we'd approach your version and tell you honestly whether AI is the right fit.