This advanced, hands-on workshop is designed for faculty and staff who already understand the basics of prompt engineering and want to explore more advanced techniques. Building on foundational skills, this session dives into crafting sophisticated prompts, optimizing AI outputs, and creating basic Microsoft Copilot large language model agents.

You’ll gain:

• Exposure to advanced prompting techniques such as few-shot, chain-of-thought, self-consistency, and failure-driven prompting
• Design patterns for clear system prompts: explicit roles, goals, boundaries, and decision rules for reliable behavior
• Agent design fundamentals, covering system prompts, pseudo-memory, and human-in-the-loop checkpoints
• Practical methods for evaluating prompt and agent reliability, including deterministic and stochastic evaluation techniques

Please note: This workshop is for faculty and staff only. Students are not eligible because institutional Copilot access is required for agent creation.

Whether you're refining workflows or building lightweight Copilot agents, this workshop equips you with the tools to move beyond simple Q&A toward structured reasoning and adaptability.

Park 207