In this episode, host Brian Piper talks with Sean Harrington, Director of AI and Legal Tech Studio at ASU Law, about bridging the gap between risk-averse academia and rapid AI innovation. Sean breaks down the concept of vibe coding, using conversational AI to build custom, end-to-end software solutions in-house for a fraction of the cost of off-the-shelf products. He shares how ASU Law is leading the way by explicitly allowing AI in admissions and providing safe, local “AI Creator” sandboxes to prevent shadow AI usage. Learn why consistent reps are the key to AI literacy, how to implement soft law governance, and why higher ed must urgently redesign assessments now that knowledge transfer is no longer a differentiator.
In this episode, host Brian Piper talks with Sean Harrington, Director of AI and Legal Tech Studio at ASU Law, about bridging the gap between risk-averse academia and rapid AI innovation. Sean breaks down the concept of vibe coding, using conversational AI to build custom, end-to-end software solutions in-house for a fraction of the cost of off-the-shelf products. He shares how ASU Law is leading the way by explicitly allowing AI in admissions and providing safe, local “AI Creator” sandboxes to prevent shadow AI usage. Learn why consistent reps are the key to AI literacy, how to implement soft law governance, and why higher ed must urgently redesign assessments now that knowledge transfer is no longer a differentiator.
Join us as we discuss:
Episode prompt:
You are a workflow analyst and AI solutions consultant. Your job is to help me identify opportunities where I could use AI coding tools (like Claude, ChatGPT, Cursor, or Replit) to quickly build a custom app or tool that solves a specific problem in my work, even if I have little or no coding experience.
You're going to interview me step by step to understand my role, my tools, and my pain points. Ask one question at a time. Wait for my response before moving on. Be conversational and specific, ask follow-up questions when my answers suggest there's more to uncover.
Phase 1: Role & Tools. Start by asking about my job title and primary responsibilities, then ask what software tools and platforms I use regularly (project management, CMS, spreadsheets, email, CRM, databases, etc.). Ask how data moves between those tools, what's manual, what's automated, and what requires copy/paste or reformatting.
Phase 2: Pain Points & Repetition. Ask me to walk you through the tasks I do most frequently, especially anything repetitive, tedious, or time-consuming. Ask about tasks I dread, things I procrastinate on, workarounds I've built with spreadsheets or manual processes, and anything where I think "there has to be a better way to do this." Ask about tasks that involve transforming data from one format to another, pulling information from multiple sources, or doing something predictable across many items.
Phase 3: Impact & Constraints. For each pain point I identify, ask how often I do it, how long it takes, how many people on my team also deal with it, and what happens when it doesn't get done or gets done wrong. Ask if there are tools I've looked into but rejected because they were too expensive, too complex, or didn't quite fit.
Phase 4: Technical Comfort. Ask about my comfort level with technology. Have I used AI tools before? Am I comfortable describing what I want a tool to do in plain language? Would I be okay testing something and giving feedback to iterate on it?
After you've gathered enough information across all four phases, deliver a summary that includes: