Cardinea helps Canadian organizations build, train, and evaluate AI on their most sensitive data. Annotation, preference data, model evaluation, and synthetic data, delivered by a cleared Canadian team on sovereign infrastructure. Your data never leaves Canadian control.
Our Approach
Your customer records, operational history, and domain knowledge are what make your AI worth building. They are also what you cannot afford to expose. Cardinea turns that data into structured, AI ready datasets inside Canadian controlled infrastructure, so you can adopt modern AI without sending your most sensitive asset to a foreign owned vendor or an offshore shop.
Every byte stays on Canadian hosted, Canadian controlled infrastructure. No foreign parent, no offshore access to sensitive data, no cross border transfer. Sovereignty is the default, not an upgrade.
Train, fine tune, and evaluate models on the data that matters most to you. We build the annotation, preference data, and evaluation sets that modern AI needs, while your underlying records stay protected.
Sensitive work is handled by a security cleared Canadian team operating under strict quality and audit controls. The people who touch your data are accountable under Canadian law.
Aligned to PIPEDA and Quebec Law 25, designed to meet Government of Canada Protected B standards, with SOC 2 Type II and ISO 27001 in progress. The controls are in the architecture, not bolted on.
Trust and Compliance
We do not just promise security, we architect it into everything we build. Canadian sovereignty, formal compliance frameworks, and a leadership team with decades of proven enterprise data protection.
Incorporated, operated, and led entirely from Canada. No foreign parent, no offshore control. Full Canadian sovereignty over every decision.
All client data resides exclusively on Canadian hosted infrastructure. Data never leaves Canadian borders.
Actively pursuing SOC 2 Type II certification to formalize the security controls already embedded in our operations from day one.
Working toward ISO 27001 certification, the international gold standard for information security management systems.
Full compliance with Canada's Personal Information Protection and Electronic Documents Act for all data handling.
Infrastructure and processes designed to meet Government of Canada Protected B security standards for sensitive information.
Process
We map the data you want to put to work, define the AI use case, and agree the access model and controls before any data moves.
Our cleared Canadian team annotates, structures, and evaluates your data inside sovereign infrastructure. Every byte stays in Canada. Every action is logged and auditable.
You receive structured, AI ready datasets and evaluation results you can use to train, fine tune, or validate your models, with full provenance and an audit trail.
Our Difference
Your organization already owns decades of unique knowledge. Cardinea turns it into structured AI assets inside Canadian controlled infrastructure, handled by cleared Canadian staff, so you can adopt modern AI without surrendering control of your data.
Incorporated, operated, and led entirely from Canada by a founder holding active Government of Canada security clearance. Every decision is made on Canadian soil.
Your most sensitive data is handled only by security cleared Canadian staff. Access, storage, and governance never leave Canadian control. Accountability sits under Canadian law, not a foreign jurisdiction.
A managed services model refined across two decades of global operations, applied to the AI data pipeline with measured quality control and full auditability.
Capabilities
End to end AI data services under full Canadian control.
Human preference labeling, ranking, and comparison data for reinforcement learning from human feedback. Delivered by domain experts under strict quality frameworks.
Multi modal annotation across text, image, audio, and video. Classification, NER, sentiment, bounding boxes, segmentation, and custom taxonomies at scale.
Red teaming, safety evaluation, accuracy benchmarking, and bias detection. Structured human evaluation protocols that give you confidence before deployment.
Privacy preserving synthetic datasets, augmentation, and scenario generation for cases where real data is scarce, sensitive, or restricted. Train and test without exposing the underlying records.
Leadership

Victoria, BC · Canada
I build the secure, data driven infrastructure that governments and enterprises can rely on, and I do it from Canada.
For two decades I have scaled technology enabled operations across three continents. As COO of TaskUs (Nasdaq: TASK), I helped scale a global organization past 20,000 people across six countries and stood up a globally managed data science discipline. As co founder of Athena, I built a roughly $100M platform in five years on a proprietary software backbone serving Airbnb, Meta, Uber, and similar clients.
My focus now is the data layer that powers modern AI, the human judgment and structured data that turn raw models into deployable systems. There is a real opportunity to build sovereign, trusted, Canadian capability in a domain currently dominated from elsewhere, and to keep the jobs, the IP, and the data onshore.
20+ Years
Scaling global operations
$100M+
Platforms built from zero
20,000+
People scaled at TaskUs
6 Countries
Global delivery experience
Adopt the next generation of AI on the knowledge your organization already owns, without surrendering control of it. Start with a focused pilot.
Explore a Pilot