7 hours ago
Vancouver, CanadaSenior
Responsibilities
- Build per-model, per-customer, and per-category accuracy reporting for Product and Sales.
- Capture customer quality-control corrections and convert them into structured, versioned training data with human review for low-confidence cases.
- Standardize deployment paths, evaluation harnesses, and model and dataset registries across restoration design, scan quality, prescription extraction, and agent squads.
- Solve multi-tenant model behavior across customer-specific product codes, prescription conventions, and preferences.
- Scale inference workloads from on-premise to the cloud with visibility into cost per inference and concurrency capacity.
- Set technical direction for AI across the company and align engineers across squads and time zones.
Requirements
- Significant experience building and operating production machine learning platforms, including model registries, versioned datasets, model CI, monitoring, and rollback.
- Depth in document and information extraction from unstructured and inconsistent text.
- Experience turning user corrections into improved models and explaining the complete data flow.
- Strong evaluation rigor, including defining and defending accuracy standards and shipping against them.
- Ability to determine when deterministic rules are preferable to machine learning.
- Ability to align engineers across squads and time zones without formal reporting authority.
- Preferred: experience with multi-tenant or per-customer model behavior at scale.
- Preferred: 3D or geometry-based machine learning experience.
- Preferred: healthcare or other regulated-data experience, including patient privacy and data residency constraints.
- Preferred: inference cost-management and on-premise-to-cloud migration experience.
Benefits
- Full-time, in-office role in Vancouver, five days per week.
- Remote and hybrid arrangements are not offered.
Categories
