Context Engineer

Royal Bank of Canada•Edina, MN

16h•Onsite

About The Position

We are embedding GenAI into everyday workflows for Financial Advisors and operational teams—starting with simple, high-value prompt use cases and expanding into agent-enabled experiences. This role is a frontline builder. As a Context Engineer (Prompt Analyst), you will focus on creating high-quality saved prompts that advisors and teams use daily and supporting agent workflows by designing and refining the prompts and context that drive them. This is a new capability for the organization. You will play a critical role in proving how GenAI can deliver real, measurable productivity gains—quickly and safely.

Requirements

Experience working with GenAI tools (e.g., ChatGPT, Claude, or similar) in a hands-on, applied way
Demonstrated ability to write clear, structured prompts
Demonstrated ability to iterate based on output quality
Strong problem-solving skills with ability to break down business tasks into steps
Comfort working with both business stakeholders (advisors, operations) and technical teams (AI engineering, data)
Basic technical fluency: Understanding of APIs, structured data (JSON), or similar concepts
Ability to work in a fast-paced, test-and-learn environment

Nice To Haves

Exposure to prompt engineering techniques (few-shot, role prompting, structured outputs)
Exposure to agent frameworks or workflow automation tools
Background in financial services or wealth management
Experience with data analysis or business analysis
Experience with user experience or workflow design

Responsibilities

Design simple, reusable prompts for internal AI tools (e.g., copilot-style usage)
Support high-frequency workflows such as: Client summaries and insights, Meeting preparation and follow-ups, Book of business reviews, Operational and service tasks
Ensure prompts are: Easy to use, Consistent in output, Grounded in the right business context
Continuously refine prompts based on user feedback and observed performance
Build and test prompts used inside agent workflows
Structure inputs and outputs for multi-step processes
Help improve how agents: Interpret instructions, Use data and tools, Generate reliable responses
Partner closely with advisors and business teams to understand: What they are trying to accomplish, Where GenAI can help (and where it should not be used)
Convert real-world workflows into: Clear prompt instructions, Structured inputs and expected outputs
Apply wealth management context (clients, accounts, portfolios, compliance considerations)
Run structured testing on prompts to evaluate: Accuracy and completeness, Consistency across runs, Risk of hallucination or misinterpretation
Identify failure patterns and refine prompts accordingly
Support lightweight evaluation approaches (prompt comparisons, A/B testing)
Help document and maintain: Prompt templates and reusable patterns, Guidance on when to use simple prompts vs. more complex solutions
Ensure prompts align with: Data privacy expectations, AI governance and risk standards