Online Session Locator

Session Description

Free ticketed event
WORKSHOP TITLE
The “Do More With Less” Cookbook: Using Generative AI to Accelerate your Day-to-Day Work as an Engineering Educator

WORKSHOP PRESENTERS
This workshop will be led by Dr. James C. Davis (davisjam@purdue.edu), Assistant Professor of Electrical and Computer Engineering at Purdue University. Dr. Davis’s expertise on this topic stems from:
- Research projects on the development and use of neural network models, funded by Google, Cisco, Socket, and the US National Science Foundation
- Integration of Generative AI into his coursework, discussed in an ASEE’24 paper in the Software Engineering Division
- Extensive use of Generative AI in his own teaching work

COLLABORATION
This workshop involves no formal collaboration with ASEE divisions.

The content of this workshop will be developed as part of NSF #2452533, on which I am Co-PI. However, that award is a research award rather than a “Workshop” award, and this workshop involves no financial sponsorship from an NSF Workshop Award.

EXPECTED AUDIENCE
The intended audience is the engineering education professoriate whose administrations are urging them to use Generative AI technologies such as ChatGPT to be more efficient, who are working with increasingly constrained resources, and who have (quite reasonably) not had as much time as they would like to optimize their use of Generative AI technologies.

LEARNING OBJECTIVES OF WORKSHOP
By the end of this workshop, participants will be able to:
• Explain the technical origins and capabilities of LLM-based Generative AI technology
• Identify tasks that are suited for partial or complete automation by Generative AI
• Apply Generative AI to complete such tasks
• Coach their colleagues and their teaching assistants in the use of Generative AI

BRIEF DESCRIPTION OF WORKSHOP
Engineering faculty are working in universities that are increasingly resource-constrained, while simultaneously facing rising expectations for responsiveness, productivity, and student support. Generative AI—particularly large language model (LLM) systems such as ChatGPT—offers real opportunities to “do more with less,” yet many educators have had little time to explore these tools or understand suitable use cases.

This workshop provides a practical, evidence-informed introduction to using LLM-based Generative AI to accelerate everyday academic work. Participants will begin by grounding their understanding in the technical foundations and capabilities of LLMs, emphasizing what these systems can and cannot reliably automate. Guided examples will then demonstrate common engineering-education tasks amenable to partial or full automation, such as drafting assignment scaffolds, preparing lectures and slides, and organizing materials. These examples draw from the workshop leader’s experience using Generative AI for these purposes in his teaching at Purdue University. Attendees will practice completing such tasks with prompting workflows they can adopt immediately.

The workshop will conclude with strategies for coaching colleagues and teaching assistants, focusing on responsible use, academic integrity, and sustainable integration into departmental practices. Participants will leave with a concrete “cookbook” of ready-to-use patterns that meaningfully reduce workload while preserving – or improving! – instructional quality.

PLANNED SCHEDULE OF WORKSHOP
(This is a 2 hour + 30 minute schedule, starting at time 0:00)
0:00–0:15 — Opening Activity: Pair-and-Share on Generative AI Experiences (15 minutes)
Participants individually reflect on their current use, hesitations, and institutional pressures regarding Generative AI. They then pair to compare experiences. The facilitator debriefs themes to surface shared challenges and establish the workshop’s relevance.

0:15–0:45 — Foundations: How Generative AI Works and How to Work with It (30 minutes)
A concise technical primer on LLM-based Generative AI (“What’s under the hood?”), focusing on capabilities, limitations, and sources of error.
Introduction to effective prompting strategies (the role of style, context, and iteration).
Brief survey of state-of-the-art tools used in engineering education.
Discussion of the value of a “tinkering mindset” in working with new technology.
Emphasis of the need for productive mistrust—engineering habits essential for safe, reliable use of AI tools.

0:45–1:10 — Activity 1: Designing a Lecture with Generative AI (25 minutes)
Small groups build a mini lecture package on a topic of their choosing (outline + 2–3 representative slides).
Participants practice specifying constraints, transferring domain expertise into prompts, and refining outputs.
Whole-room sharing of strategies and pitfalls.

1:10–1:35 — Activity 2: Triaging Student Assignments (25 minutes)
Participants work with anonymized or synthetic student submissions to practice:
• generating formative feedback,
• identifying common misconceptions,
• organizing responses efficiently.
Discussion centers on boundaries of responsible automation and preserving instructor voice.

1:35–2:00 — Activity 3: Troubleshooting, Debugging, and Role-Play Coaching (25 minutes)
Participants rotate through brief troubleshooting scenarios (e.g., misleading outputs, missing context, incorrect citations).
In pairs, they practice coaching a teaching assistant on how to prompt effectively and how to check AI-generated work for quality and integrity.
Debrief emphasizes departmental sustainability and scalable practices.

2:00–2:30 — Integration, Cookbook Patterns, and Closing Discussion (30 minutes)
The facilitator synthesizes the workshop into reusable prompting “recipes” for common academic tasks.
Participants identify 1–2 workflows they will adopt immediately and share implementation plans.
Q&A and closing reflections on long-term, responsible use of Generative AI in engineering education.

FUNDING SOURCE OF WORKSHOP MATERIAL
The content of this workshop is developed as part of NSF #2452533.

WORKSHOP TIME
No time restriction – morning or afternoon is fine.

Speaker

1. James Davis

   Purdue University at West Lafayette (COE)

   James C. Davis is a sixth-year Assistant Professor in the Elmore Family School of Electrical and Computer Engineering at Purdue University. He studies the engineering of computing systems, writ broad, with a focus on the effective use and re-use of machine learning models. His primary interests are in software correctness, security, and usability. He has published over 70 peer-reviewed papers, including 40 research papers at prestigious software engineering and cybersecurity venues. He is the inventor of 8 US patents. His work has received 3 ACM Distinguished Paper Awards and 1 ASEE Best Paper award (software engineering division). His research group has been funded by the US National Science Foundation and by Google, Cisco, Rolls Royce, Qualcomm, and Socket. He has received departmental, college-level, and university-level awards for his teaching and mentoring.