This page last updated:
03.06.26 3:50 pm
Welcome! This page showcases my instructional design work for the University of Florida College of Journalism and Communications hiring team. Thank you for taking the time to peruse this work.
About Andy Sheppard
As detailed on my resume, I worked 17 years at the University of Florida and 10 years at Santa Fe College, providing instructional design support for both in-person teaching and online learning. Moreover, I’ve spent the past 12 years engaged in entrepreneurial pursuits centered around media development.
The page linked below includes prior work done at Santa Fe College. My colleagues were kind enough to leave this course publicly accessible. Please note that I did need to remove some of the videos from the Teaching Examples module to meet newer SF brand guidelines. I also no longer have direct editing access, so I’m not able to further update this Canvas course site. This portfolio supported my application for my last post at SF, and thus, was written with an SF audience in mind.
Santa Fe College Work Examples 
In the past couple of years, I’ve largely focused on designing training and creating media using emerging Artificial Intelligence (AI) tools. As the following illustration suggests, I’m passionate about the intersection of AI tools, pedagogy and digital accessibility.
Above: An AI-generated Venn diagram.
Attraction to the UF College of Journalism and Communications
As noted on my resume, I earned a Minor in Mass Communication at UF, and journalism remains a strong interest of mine—especially as social media competes with so-called legacy media for attention. Ultimately, one “fact error” on an assignment prompted me to change my educational plan. 💁🏻♂️ At the time, a single fact error equated to a 50% reduction in the assignment grade. Many years later, I feel equipped to have a robust collegial discussion about this type of well-intended but perhaps not fully-aligned course policy.
More recently, a colleague told me something that has stuck with me. After revamping a Media Studies course, he said his students saw news differently than we might expect: to them, it’s not a distinct category of media. A video from ABC News might appear in the same scroll as an influencer clip, a cat video, or a podcast segment. From their perspective, it’s all the same—“just content.” I’m eager to collaborate with faculty to help students understand the important distinctions between respected news organizations and some random individual posting content online.
Work Examples
Academic Example – Artificial Intelligence Training
I scripted and edited the following video for SF College’s ‘AI Toolkit,’ introducing recent AI advances to faculty. I worked collaboratively with the campus’s AI committee to design this training. My goal was to “show, not tell”. While our studio staff can edit our videos for us, I prefer to keep my video editing skills fresh. It’s also quicker for me to produce a video that I’ve personally story-boarded. This video was created prior to Open AI’s and Google’s recent advances in image generation.
Design reflection: Ideally, I would feature faculty in a video like the one above. In practice, however, scheduling recordings when both the studio and faculty are available has proven logistically difficult, and many faculty are understandably hesitant to be recorded.
Academic Example – An Unusual Introduction
Last year, I applied for a position with UF’s AI Learning Academy and wanted my application to stand out. I reached out to a couple of voice actors I’ve worked with and wrote the opening of a script, inviting them to adapt it and share any concerns they had about AI in their field. I didn’t expect the video to take such an unexpected—but important—turn. Erin’s insights, in particular, were profound and sobering.
After reflecting on Erin’s commentary, I decided to check out Google’s Notebook LM tool. It converts documents and other files into a simulated podcast. And, my first LM-produced podcast supports her concerns.
I produced a handful more podcasts with the tool, and while this is only anecdotal, the male typically leads the podcast, in my experience. Others have raised concerns about this bias. More broadly, Dr. Joy Buolamwini has been warning us about the problems with bias in AI technology for years. Every educator should probably be aware of the Algorithmic Justice League.
These biases pose a key pedagogical challenge for journalism faculty: how can we equip students to critically assess the tools they are likely to use in professional practice?
Academic Example – Gamified Tutorial
The following multimedia experience is a Spider-Man themed interactive Canvas tutorial, and it includes a game designed to promote growth mindset. My audience for this tutorial was first-year college students; gamification can be especially engaging for these younger learners. I designed this game using Articulate Storyline 360. Design reflection: As I look back on this work with literally different lenses – reading glasses – I realize that the text size is too small in some places.
Play Game (new window)
Commercial Example – Video Avatar Use
A client is piloting the use of AI-generated video avatars in select learning modules. We will analyze survey data to compare learner responses between modules with and without the avatars. Here’s a short excerpt from one of the instructional videos. Meet “Brandon”.
Pedagogically, I have mixed feelings about the use of AI-generated video avatars – especially within the context of higher education. These types of avatars run the risk of increasing social distance between students and their instructors. Technically, we’re still in uncanny valley, and the syncing between audio and video is a bit off. And, the cognitive dissonance of transitioning from a video featuring voice actors to AI-generated voices is not lost on me.
Design Process
While many of the instructional design examples listed on this page demonstrate applied technical skills, the more important skill-set I bring is my instructional design approach. I seek to understand the needs of the audience and focus on the knowledge or skills participants will need to gain. As Grant Wiggins and Jay McTigue articulate, this approach is known as backwards design. I design trainings and resources with the end goal in mind; and I strive to remain current on evidence-based design practices.
During the design process, I’m passionate about universal design—creating with disabled users in mind. I prioritize accessible colors, include captions and transcripts, and ensure consistent, predictable formatting. When appropriate for the audience, I make occasional use of pop culture references, memes and emoji.
After designing a training session or multimedia asset, I find a way to assess its impact. This assessment might involve the use of surveys, focus groups or tracking data from within a learning portal. For example: in my prior teaching, I would host my own instructor-created videos on YouTube and embed them within our course platform. In doing so, I could access YouTube’s more robust analytics to determine what portion(s) of videos my students would watch, and when they were typically viewed. This data would then inform my revision of the instructional content.
What Lies Ahead …
I’ve touched on a couple of the ethical concerns we’re facing regarding AI adoption; and to be sure, there are many more to consider. However, at the end of the day, these tools are increasingly baked into our operating systems and devices. For example, the more recent version of Microsoft Paint has an image creator built within it. Instructional Designers are at their best when they work collaboratively with their colleagues to critically evaluate emerging educational technology tools.
Closing Thoughts
I’ll close with a video I scripted and produced in summer 2024. When I proposed adding it to the SF AI Committee’s AI Toolkit, it sparked debate, but most agreed it addressed an emerging issue. Recently, AI companions have been linked to troubling teen suicide cases, and as more people seek therapy or guidance from chatbots, concerns are growing about how these systems develop evolving “personalities” and relationships with users.
From a journalism perspective, these developments raise important questions about audience trust, transparency, and the future of mediated communication. One can imagine a near future in which audiences select a realistic AI avatar to deliver the news based on their individual preferences—blurring the line between journalism, personalization, and synthetic media.
Thank you for considering my work examples. I appreciate your time and hope you have a great day!
astermedia.org 