Agent Lead

RBC•Minneapolis, MN

3d•Onsite

About The Position

The Agent Lead sits at the forefront of transforming Financial Advisor productivity through agentic AI workflows—bridging business problems with intelligent automation. This role operates in a high-ambiguity, rapid experimentation environment, identifying where agent-based approaches can unlock value—and equally, where they should not be applied. You will define how vendor agents (e.g., CRM-native), enterprise frameworks, and internally developed agents coexist and interoperate within a governed ecosystem. This is not a pure engineering role—it is a product-minded builder who shapes, validates, and scales agentic patterns that can be reused across Wealth Management. The mandate is to move fast, prove value, and establish repeatable patterns, while aligning to enterprise architecture, risk, and AI governance standards.

Requirements

Proven experience building and deploying agentic AI solutions (multi-agent systems, orchestration frameworks, tool-using agents)
Strong understanding of agent frameworks and architectures (e.g., orchestration layers, memory models, tool integration, event-driven agents)
Demonstrated ability to operate as a builder + product owner hybrid
Experience working across business, engineering, and enterprise governance functions
Ability to rapidly prototype (POCs) and iterate based on real user feedback
Strong judgment in when to use AI vs. when not to
Experience designing workflow-driven automation (not just models)
Leadership experience managing technical talent (e.g., prompt/context engineers)
Excellent stakeholder engagement skills—comfortable working “side-of-desk” with advisors and executives

Nice To Haves

Experience in Wealth Management / Financial Services, particularly advisor workflows
Familiarity with CRM-based agent platforms (e.g., Salesforce Agentforce)
Exposure to event-driven architectures and real-time data integration
Understanding of AI risk, model governance, and explainability frameworks
Experience integrating with enterprise AI platforms (e.g., internal AI platforms, cloud AI services)
Background in human-centered design or workflow optimization

Responsibilities

Agentic Strategy & Use Case Qualification: Partner directly with Financial Advisors, field leadership, and business stakeholders (“side-of-desk”) to identify high-value workflow opportunities. Evaluate when to apply agentic AI vs. deterministic automation vs. no automation, with the authority to say “this is not an AI problem”. Define and prioritize agentic use cases aligned to advisor productivity, client engagement, and operational efficiency.
Agent Design, Build & Rapid Experimentation: Lead rapid POC development cycles (fail fast / scale fast) for agentic workflows. Design multi-agent interactions across: Vendor agents (e.g., CRM/Agentforce), Enterprise agents (shared services / platforms), Native/internal agents (event-driven, workflow-specific). Establish reusable agent design patterns (prompting, orchestration, memory, tool usage, escalation paths). Partner with AI Engineering to validate feasibility, performance, and scalability.
Product Ownership & Lifecycle Accountability: Act as Product Owner for agentic workflows—owning use case shaping through validated solution patterns. Ensure solutions are not “built and dropped” by: Defining success metrics and adoption criteria. Driving iteration based on advisor feedback and usage telemetry. Maintain a portfolio of agentic capabilities with clear value articulation and reuse potential.
Enterprise Alignment & Governance Integration: Engage with enterprise stakeholders (e.g., Borealis / enterprise AI, architecture, and platform teams) to: Align with approved agentic frameworks and standards. Leverage existing enterprise capabilities before building net new. Ensure all agentic solutions align to: Model risk governance, Data privacy and security requirements, AI explainability and control frameworks.
Agent Ecosystem & Standards Definition: Define how agent ecosystems operate within RBC Wealth Management, including: Interaction models between vendor, enterprise, and native agents. Guardrails for agent autonomy and decisioning. Cost-efficiency and performance considerations. Contribute to evolving enterprise agent standards through applied learnings.
People Leadership & Capability Building: Manage and develop Context Engineers / Prompt Engineers. Establish best practices in: Context design and retrieval strategies, Prompt engineering and agent behavior tuning. Build a culture of experimentation, accountability, and pragmatic problem solving.
Field Engagement & Adoption: Travel (~25%) to branches and field locations to: Observe advisor workflows firsthand. Identify friction points and real-world opportunities for agents. Validate usability and adoption of agentic solutions.