Medical-Imaging Algorithm GCC Quality Leader

The mandate

A production algorithm update improved aggregate sensitivity while degrading performance in a smaller patient subgroup not represented in the final regression set. India engineers implemented and tested the change, but Toronto scientists retained cohort sufficiency and limitation judgments outside the formal workflow. The algorithm-quality leader exited after quality review found two more products with the same decision gap.

The interim must start in Toronto within three weeks and lead for fourteen months across Canadian and India teams. Search for a permanent algorithm-quality executive opens after two affected products complete subgroup evidence reviews and one changed-distribution simulation, expected by month eight. The successor will chair a limitation decision and corrective-action review during six weeks of overlap.

Handover requires five algorithm products to have intended-use and cohort evidence, traceable change classifications, subgroup performance thresholds, defined quality and clinical interfaces, exercised rollback and India quality leadership depth. Three distribution-shift or data-quality scenarios must be witnessed. The successor inherits limitation registers, corrective actions, evidence gaps, source dependencies and talent commitments.

The interim may stop an algorithm release, require expanded evidence, classify a change for heightened review, redirect quality-engineering capacity and commit up to CAD 18 million within the approved remediation portfolio. Medical use, independent quality approval, regulatory submission, market withdrawal and clinical claims remain with authorised functions. India leaders gain algorithm-quality recommendation and evidence-acceptance rights under documented thresholds.

Clinical practice, enterprise imaging-platform replacement, data acquisition, legal interpretation and products outside the five named algorithms are excluded. The seat covers engineering quality, cohort and change evidence, corrective action, India leadership and succession. It cannot act as independent approver or substitute aggregate metrics for accountable clinical and quality judgment.

Why this seat is open

The subgroup degradation revealed that code and aggregate regression had transferred while consequential evidence judgment had not, followed by leadership exit. Quality, clinical and engineering teams now disagree over how much remediation is required. A temporary imaging-algorithm leader can restore product-level evidence and exercise India authority before permanent succession.

What you will own

• Reconstruct the update across intended use, cohort composition, performance measures, subgroup result, change decision and release evidence.
• Define algorithm-quality authority over test sufficiency, change classification, limitation recording, rollback recommendation and corrective action.
• Preserve independent quality and medical approval while moving routine evidence judgment into India.
• Establish dataset, annotation, provenance, representativeness and leakage requirements for each named product.
• Build India algorithm-quality, imaging-science and verification leaders with exercised deputy and clinical-review exposure.
• Run distribution shift, device variation, annotation disagreement, subgroup degradation and rollback scenarios.
• Transfer product evidence, limitation histories, corrective actions, leadership decisions and unresolved quality risks to the successor.

Candidate qualifications

• Held senior quality or product authority for regulated medical-imaging algorithms across development, deployment and post-release change.
• Recovered clinically relevant subgroup or distribution weakness hidden by otherwise acceptable aggregate validation measures.
• Built India algorithm-quality leaders trusted by imaging scientists, clinicians, independent quality officers and product executives.
• Distinguished engineering evidence acceptance from accountable medical judgment, independent approval and regulatory submission responsibility.
• Led corrective action across cohort design, annotation, algorithm change, regression evidence and limitation communication.
• Handed a regulated algorithm portfolio to permanent leadership after observed release, rollback and corrective-quality decisions.

Non-negotiables

• Can begin in Toronto within three weeks and travel monthly between Canada and India algorithm teams.
• Will accept exclusive executive accountability and urgent escalation for the five algorithm-quality products.
• Brings regulated imaging-algorithm quality plus India capability transfer; general AI testing is insufficient.
• Must disclose medtech, imaging-provider, data, annotation, clinical-research and algorithm-vendor relationships.

1. 49 words maximum. Describe an algorithm update whose aggregate result concealed clinically important subgroup degradation.
2. 49 words maximum. Which evidence judgment can India own without taking independent quality approval?
3. 49 words maximum. State your Toronto availability and the largest regulated algorithm portfolio you directly governed.