AUDIENCE
Intermediate learners with a working knowledge of programming fundamentals (variables, functions, and JSON handling)
DATE
Apr 27, 2025
SMEs
Hardeep Johar
Teaching Professor in the Department of Industrial Engineering and Operations Research
TASKS
PROBLEM STATEMENT
Understanding how to work with large language models (LLMs) is becoming essential, as natural‑language interfaces are rapidly integrating into everyday products, from search engines to office software. Professionals who can craft clear prompts, navigate model limitations, and safely embed AI into workflows are in high demand. Mastering these skills not only future‑proofs your career but also gives your organization a competitive edge by accelerating development cycles and enabling new kinds of user experiences.
THE PROCESS
The course was developed through an iterative design process that combined instructional design frameworks with hands-on prototyping. The team began by mapping learning objectives to real-world AI use cases, then built and tested prompt engineering exercises and API integration labs in parallel to ensure a seamless learning flow. Each module went through rounds of faculty review, learner testing, and technical QA to optimize clarity, pacing, and accessibility.
Over the process of developing this course, I've gathered many key lessons for my work as an ID:
The need to balance conceptual and technical depth: learners benefit most when theory directly connects to coding practice.
Prompt engineering is best learned experientially, through iterative design and reflection rather than static examples.
Designing for an asynchronous audience requires intentional scaffolding, clear guidance, and lightweight interactive elements to sustain engagement.
Close collaboration between subject matter experts, instructional designers, and developers is critical to align pedagogical intent with technical implementation.
A major design challenge was translating complex technical concepts, such as model parameters, tokens, or API calls, into approachable, real-world explanations without oversimplifying. This required blending narrative examples, visual diagrams, and hands-on code exploration to make abstract ideas concrete for diverse learners. This experience deepened my appreciation for the skill of bridging complexity and clarity, transforming advanced AI concepts into learning experiences that invite curiosity and empower non-experts to explore, build, and create confidently.
FINAL PRODUCT
This course culminates in the development of a working mini-application that integrates OpenAI’s API to perform real-world generative AI tasks such as text generation, summarization, or question answering. By completing this capstone project, learners demonstrate both prompt engineering proficiency and hands-on programming skills.
Throughout the process, learners will:
Design and refine prompts for targeted use cases.
Connect to and interact with the OpenAI API using Python.
Build, test, and document a functional prototype such as a chatbot, content generator, or intelligent assistant.
The final deliverable showcases each learner’s ability to translate prompt design into practical implementation, serving as both a portfolio piece and a foundation for further innovation in generative AI applications.
