This advanced course is designed for professionals who already understand the fundamentals of AI prompt engineering and want to move beyond basic prompting into more precise prompt design and reasoning control.
Building on foundational prompt engineering concepts, this course focuses on how experienced users design prompts intentionally, not just to generate responses, but to guide reasoning, surface uncertainty, reduce errors, and improve reliability in complex professional use cases.
Participants will learn how to design multi-prompt systems, apply self-critique and adversarial prompting, manage ambiguity and uncertainty, and evaluate AI outputs using comparative techniques. The emphasis is on advanced cognitive strategies rather than specific tools or automation, making the course broadly applicable across roles and industries.
What You Will Learn
- Design multi-prompt systems that break complex tasks into controlled reasoning stages.
- Apply self-critique and adversarial prompting to identify weaknesses and reduce AI errors.
- Use constraint-first prompting to control scope, exclusions, and acceptable outputs.
- Prompt AI systems to surface uncertainty rather than guess or hallucinate.
- Evaluate AI outputs using comparative prompting and confidence calibration techniques.
- Optimize prompts for clarity, efficiency, and repeatability without sacrificing quality.
Course Outline
Module 1: Prompt Architecture & Multi-Prompt Systems
-
Explain the difference between single-prompt tasks and multi-prompt reasoning systems.
-
Decompose complex work into discrete prompt roles.
-
Design a coordinated set of prompts that each serve a specific reasoning function.
Module 2: Reasoning Control Through Self-Critique & Adversarial Prompting
-
Use self-critique prompts to identify weaknesses in AI outputs.
-
Apply adversarial prompting to surface gaps, risks, and assumptions.
-
Improve output quality without adding new source material.
Module 3: Constraint-First Prompting & Uncertainty Control
-
Design prompts that explicitly control scope, exclusions, and failure conditions.
-
Instruct AI to surface uncertainty instead of guessing.
-
Reduce hallucinations by shaping acceptable behavior in advance.
Module 4: Comparative Prompting & Prompt Optimization
-
Use comparative prompting to identify weaknesses and inconsistencies.
-
Evaluate outputs by analyzing disagreement across prompts.
-
Optimize prompts for efficiency without degrading output quality.
This course is available for "remote" learning and will be available to anyone with access to an internet device with a microphone (this includes most models of computers, tablets). Classes will take place with a "Live" instructor at the date/times listed below.
Upon registration, the instructor will send along additional information about how to log-on and participate in the class.
