Join EY and help shape the future with confidence as an AI Prompt Engineer focused on Microsoft Copilot and Copilot Studio. This role is integral to our GDS Assurance Digital team, driving innovative AI solutions that enhance audit quality, accelerate timelines, and boost productivity.

You will collaborate with diverse, multi-disciplinary teams across Assurance, Generative AI, and platform engineering. Your responsibilities will span from translating complex Assurance use cases into intuitive conversational experiences to building, testing, and governing reliable and explainable AI prompts and agents.

We seek individuals adept at balancing value, quality, risk, and compliance, ensuring our AI solutions are traceable and continuously improved through user feedback and rigorous testing.

Design, refine, and optimize prompts (system prompts, grounding instructions, response constraints) for advanced AI interactions. Engineer multi-turn conversational flows with robust guardrails, employing techniques like role-based prompting, structured reasoning, RAG, and hallucination reduction. Build and configure Microsoft Copilot Studio agents, including topics, triggers, entities, actions, and orchestration logic, to create seamless end-user experiences. Integrate agents with enterprise data sources, APIs, and workflows, ensuring adherence to security, access control, and platform standards. Execute comprehensive AI testing strategies, including prompt validation, edge-case testing, and quality checks, to enhance accuracy and consistency.

A strong academic foundation, including a Bachelor's or Master's degree in Computer Science, Data Science, Engineering, Information Systems, Business Analytics, or a related field, with a specialization in AI-enabled product development and automation. Possess 4-7 years of overall engineering experience, with 2-3 years specifically in Generative AI, focusing on AI-enabled product roles, automation, or digital transformation delivery. Demonstrate hands-on experience with prompt engineering for LLM-based systems, building solutions using Microsoft Copilot, Copilot Studio, Python, and LangGraph. Exhibit a solid understanding of LLM behavior, limitations, risks, and evaluation techniques. Successfully translate complex business or regulatory requirements into structured AI interactions, including conversation design, intent flows, and guardrails. Experience working within Agile delivery models and collaborating with cross-functional global teams is essential.