Requisition ID: 267929
At Tangerine, we’re redefining banking. As Canada’s leading digital bank, we thrive on innovation, agility, and bold thinking, tackling every challenge head-on with leading technology and the unstoppable power of collaboration.
Our client-obsessed teams deliver flexible, accessible banking solutions, breakthrough products, and award-winning service. Beyond banking, we’re committed to making an impact in the communities we serve and across our organization. We foster a culture built on integrity, inclusion, and fearless ambition – where diverse perspectives are valued, and people are empowered to do their best work.
Are you ready to disrupt the status quo? Do you crave challenges that push boundaries? If you’re a high performer looking to accelerate your career and reimagine the banking landscape, this is your moment.
Let’s shape the future of banking together!
Is this role right for you?
- Strategy & Responsible AI Governance – Work with the business to define and implement a strategy for trustworthy, personalized natural language and agentic interfaces - embedding fairness, transparency, and accountability from design through deployment
- Act as lead engineering partner to translate business requirements into responsible architectural patterns and best practices for agentic AI development, balancing innovation with risk and ethical guardrails
- Partner with Data, security, risk, legal, and engineering teams to embed AI-driven resiliency, observability, bias detection, and threat modeling into architecture and governance standards
- Champion model risk management practices (e.g., SR 11-7), regulatory alignment, and readiness for emerging AI regulation across the agentic AI portfolio
- Architecting a Responsible Agentic Development Platform and Practice
- Work as part of a team defining and delivering the strategy, technical components, and guardrail patterns for a responsible agentic AI ecosystem
- Design and implement secure, resilient, and compliant AI platforms that integrate with existing enterprise systems, ensuring alignment with risk, ethics, fairness, and regulatory frameworks
- Build on knowledge of event-driven architecture, message queues, and distributed systems to build resilient, auditable, and scalable agentic workflows in a secure and compliant manner. – Drive search and retrieval architecture, including vector-based search systems, indexing strategies, and relevance optimization, with attention to provenance, grounding, and hallucination mitigation
- Design and embed responsible AI controls - guardrails, content and prompt safety, policy enforcement, explainability, and audit trails - directly into the platform
- Manage and iterate on new platform capabilities while balancing the needs of key projects, delivering maintainable, scalable, testable, and easy-to-refactor solutions
- Work hands-on and de-risk the critical parts of our complex projects & Evaluation pipelines. The role will be deeply involved in coding, reviewing, red-teaming, and optimizing AI solutions
- Mentor and Drive a Responsible Engineering Culture
- Inspire and mentor teams across architecture and engineering disciplines, cultivating a culture of innovation, responsibility, and trust in AI adoption
- Drive next-generation testing, evaluation, red-teaming, and observability practices for dynamic orchestration of user journeys
- Establish responsible AI review and evaluation practices - fairness testing, bias audits, safety evaluations, and ongoing monitoring - as a standard part of the delivery lifecycle
- Champion a culture of excellence, inclusion, and continuous learning within the team. When something breaks, you're not satisfied with a workaround. You dig until you understand why, and you bring learning and long-term solutions back to the team.
Do you have the skills that will enable you to succeed in this role? We'd love to work with you if you have:
- Proven experience in AI leadership and delivery roles, with a track record of successfully leading and delivering AI projects - scoping ambiguous problems, communicating trade-offs, and iterating based on real-world feedback
- Demonstrated experience embedding Responsible AI principles (fairness, transparency, accountability, privacy, and safety) into production AI/ML systems
- Extensive experience in Python and its core data science libraries (e.g., Scikit-learn, Pandas, NumPy, Matplotlib/Seaborn)
- Hands-on experience building and taking LLM-powered applications to production (retrieval, agents, structured outputs, prompt safety) with appropriate guardrails and evaluations
- Experience with Agentic AI frameworks and designing multi-step AI reasoning processes, including safety, guardrail, and human-in-the-loop design
- Experience with MLOps principles and tools for model versioning (e.g., Git), containerization (e.g., Docker), and continuous integration/continuous deployment (CI/CD) of machine learning models
- Familiarity with model risk management and AI governance frameworks (e.g., SR 11-7, NIST AI RMF, EU AI Act) and the ability to operationalize them in engineering practice
- Proven experience in automated testing and the ability to develop test strategies and design automation frameworks, including evaluation and red-teaming for AI systems
- Experience with running benchmarks at scale for model development and evaluation
- Strong experience in full-stack fundamentals, system design, and integrations, with demonstrated leadership in designing and managing low-latency microservices at scale with authentication and authorization (OAuth 2.0, OpenID Connect, and SAML).
Preferred Qualifications
- Strong theoretical and practical knowledge of classical machine learning algorithms (e.g., classification, regression, clustering, dimensionality reduction) and their applications in areas such as fraud detection, credit risk scoring, or customer segmentation
- Experience with explainability and interpretability techniques (e.g., SHAP, LIME) and bias/fairness assessment tooling
- Experience in fine-tuning LLMs, embedding models, and tackling advanced RAG or agentic use cases. Building pipelines and eval sets to benchmark performance of such solutions is essential.Demonstrated expertise in feature engineering, feature selection, and data transformation techniques for structured and unstructured data – Experience in designing AI observability, monitoring systems, feedback loops, drift detection, and trend analysis
- Familiarity with Google's Vertex AI tech stack
- Experience building applications with modern web component frameworks (such as React & Angular)
- Experience with containerized tooling and cloud platforms (particularly Docker/Docker Compose, Kubernetes, and GCP)
Location(s): Canada : Ontario : Toronto
At Tangerine we value the unique skills and experiences each individual brings to the team, and are committed to creating and maintaining an inclusive and accessible environment. If you require accommodation during the recruitment and selection process, please let our Recruitment team know.
