Ahmad Ghabboun built a Demo Day–winning AI product during his MSIS program — after arriving with no plans to work in AI at all. He breaks down how his mindset shifted, how his design background made him a stronger prompter, and how to build AI fluency that actually holds up in interviews. Useful for students and early-career professionals trying to get AI-ready without faking it.
Ahmad Ghabboun is a Master of Science in Information Systems (MSIS) 2026 Graduate at the UW Foster School of Business. Before Foster, he spent roughly fifteen years in UX and product design, building web applications for startups. At Foster he built several generative-AI tools in his coursework, including Synapse, which won Best Business and Tech Product at the MSIS Demo Day. He is targeting product management and technical product roles.
What you’ll learn
- Why naming the specific AI model you use — and justifying it — matters more in interviews than saying “I use AI”
- How a design background translates into sharper, more technical prompts
- How to keep a human in the loop so AI assists your judgment instead of replacing it
- Why AI’s tendency to agree with you makes human and second-model pushback essential
- How to stay current with fast-moving tools without trying to learn everything
- The difference between a productivity mindset and a learning mindset in school
Key moments
- The third-quarter AI classes that moved AI from “not on my list” to his career focus
- The origin of Synapse: manually juggling answers across Gemini, Claude, and a third model
- How Synapse runs a dual-model validation and a judge step to flag gaps for technical PMs
- Why interview proctoring now detects AI use — and what a “perfect” AI answer signals to interviewers
- Ethan Mollick’s “jagged edge” and why it shifts with every model release
Resources mentioned
Lovable; Replit; Gemini; Claude; ChatGPT; Jira; Azure DevOps; GitHub; Ethan Mollick’s “jagged frontier” of AI capability.