Director, AI Platform Engineering

BMO; Toronto; On-site; CAD 180,000 - 240,000; Full time

Job summary

BMO is assembling an Enterprise AI Platform Engineering function to build and operate the control plane and governance layers that enable safe, auditable AI workloads across our business domains. You will own the platform runtime, enforce policy, and drive observability across cloud environments.

This is a hands-on, build-and-run leadership role requiring collaboration with Security, Architecture, and DevOps teams to scale capabilities and ensure regulatory defensibility from day one.

Qualifications

• Proven experience leading large-scale AI platform initiatives.
• Strong background in secure, compliant platform design for enterprise AI.
• Ability to partner with Security, Architecture, DevOps and domain teams.

Responsibilities

• Own the Enterprise Control Plane and its governance tooling.
• Lead Domain Orchestration workloads across multi-cloud environments.
• Deliver measurable lifecycle, certification, and audit capabilities for AI workloads.
• Scale the platform team through hiring and internal talent development.

Skills

Platform engineering

Leadership

Security & Governance

Cloud architectures

Tools

Cedar/OPA

GitOps

OpenTelemetry

Job description

Director, Enterprise AI Platform Engineering

BMO is building a dedicated AI Engineering function to deliver the platform capabilities that make enterprise AI safe, governed, and scalable across our business domains and regulatory regimes. We are seeking an experienced technical leader to own the core infrastructure that governs and enforces how AI runs at BMO - the AI Gateway, Policy Engine, Identity Fabric, AI Registry, Guardrails Runtime, and AI Observability.

This is a build-and-run leadership role. You will lead a team that designs, ships, and operates the control and orchestration infrastructure sitting between policy authoring and inline enforcement - the capabilities every AI workload at BMO consumes to be secure, compliant, and observable. You will not own the AI models or applications themselves (those are domain-owned); you own the governed platform they run on, and the runtime evidence that proves they run within policy.

You are a hands-on technical leader who has built platform capabilities at scale, operates what you build, and designs for operability and regulatory defensibility from day one. You blend deep engineering credibility with the executive presence to partner across Security, Architecture, DevOps, and domain teams. You are energized by taking real engineering assets - an existing developer portal, AI registry, a body of policy-as-code, and gateway integrations - and formalizing, scaling, and governing them into an enterprise-grade platform.

What You'll Own

Enterprise Control Plane

• Developer Portal & AI Registry - productionize the developer portal; deliver a federated AI Registry spanning agents, models, tools, channels, and evaluations, with self-service onboarding and lifecycle workflows.
• Policy Engine - policy-as-code infrastructure (Cedar/OPA), a policy compilation and GitOps distribution pipeline, risk-tiered approval workflows, and a policy simulation environment.
• Observability & Audit - a multi-pipeline architecture spanning operational, security, and compliance telemetry; OpenTelemetry GenAI conventions; cross-pipeline trace correlation; and a tamper-evident audit lake producing regulator-ready evidence.
• Governance & Lifecycle - certification workflows, automated compliance scoring, decommission governance, and evidence generation for architecture and model-risk review.

Domain Orchestration

• Gateway Runtime - domain-hub deployment across multiple clouds; an inline enforcement engine with request-time policy evaluation, routing, residency, budget/quota controls, and circuit breakers, operating within strict latency budgets.
• Guardrails Runtime - a multi-stage safety pipeline (input moderation, prompt-injection defense, PII handling, output validation, hallucination detection, policy enforcement) with bilingual (EN/FR) parity and behavioral guardrails for agentic workloads.
• Identity Fabric - workload identity for AI (SPIFFE/SPIRE), token-exchange bridging, per-domain trust boundaries, enterprise identity integration, and cross-cloud token federation with zero-trust attestation.

What You'll Deliver (First 12 Months)

• A production-hardened Developer Portal and federated AI Registry with sub-5-day self-service onboarding.
• An AI Gateway operational in a selected business domain, meeting tiered latency targets.
• Policy-as-code infrastructure distributing domain-scoped policy bundles via GitOps, with a working simulation sandbox.
• A runtime evidence pipeline producing lineage-stamped, audit-ready traces aligned to model-risk and regulatory expectations.
• A team scaled from an initial core (8-12 FTE) toward steady-state through a blend of net-new hiring and reallocation of experienced internal engineers.

How You'll Work

• Build-run integrated - your team operates what it builds; there is no separate run team. You design for operability and Engineering support from the start.
• Federated - you own the enforcement infrastructure domains consume; domains own their workloads. You enable, you don't centralize execution.
• Evidence-first - regulatory evidence (e.g., OSFI E-23, OCC model-risk expectations) is produced at runtime through instrumented infrastructure, not assembled retroactively.
• Capability-aligned - your teams own outcomes (\