Amid the chaos of a world in crisis, I’ve found hope in an unexpected place: coding. With tools like Claude.ai and MCP, I’ve been building a web app to help food pantries serve their communities better—automating inventory, breaking language barriers, and streamlining processes. This isn’t just about code; it’s about turning anxiety into action, using technology to create something meaningful. If you’ve ever wondered how AI can amplify human effort, this is a story for you.
Read MoreDesign For Service - Case Study
Case Study: Frankie Bunz, Pittsburgh, PA
I moved to Pittsburgh in August, 2019. Since moving here, I have eaten at only a handful of restaurants; Frankie Bunz (i.e., mobster-themed hot dogs) is easily one of my favorite local places to grab a bite. It is in Squirrel Hill, on Murray Avenue.
While this restaurant does offer some dine-in seating, it is primarily designed for grabbing food to go. In evaluating their services, I opted to dine in.
Customer Journey: Phase 1 - discovery
Customers are most likely to be attracted to this restaurant if they are on foot. The sidewalk immediately outside of Frankie Bunz advertises daily specials. There is a full menu in the window, as well as flyers promoting their most recently added items.
Customer Journey: Phase 2 - Entry
When you walk into Frankie Bunz, it immediately becomes clear that they do not have a large seating capacity, but they still provide an inviting atmosphere. Additionally, they provide a large banner-type version of their menu.
Customer Journey: Phase 3 - Ordering
I arrived for a late lunch (this first week of the semester has started out with many plates for me to spin, including this evaluation), and the only other customers were take-out or app-based delivery workers (e.g., Grubhub). The ordering and checkout process is reasonably frictionless. They use a touchscreen POS machine with contactless (Apple, Google, Samsung, etc.) and chip-reading capabilities.
Customers can either choose one of the standardized hotdogs, or build their own. The staff takes the order, unless the customer is ordering via a delivery app. Customers ordering a standardized hotdog (e.g., “Fredo’s Frank” or “The Don”) are still asked what kind of bun they’d like. Options include: wheat, white, pretzel, and onion roll. In addition to their buns, they also offer a tempura battered, fried dog on a stick (i.e., a “corndog” minus the cornmeal); they call it the “Mr Miyagi Doggie” and it includes an Asian Fusion slaw and special “dragon sauce.”
Customer Journey: Phase 4 - Payment
The only substantial flaw with this setup is the counterintuitive chip-reader.
The arrow is pointing away from the slot, but this doesn’t necessarily clarify card orientation. The affordances of the device allow for both correct and incorrect insertion. In total, this card-reading device allows no less than eight card orientations and interactions (four in the card slot, and four in the slider), and only one of these actions is correct. To be generous, there is at least an 87.5% chance for error, even with written instructions. This is terrible design.
Despite this minor annoyance, the process is still supported by staff, and any errors can be quickly observed and corrected.
Customer Journey: Phase 5 - Fulfillment
Once the order is placed and the payment confirmed, customers have a brief waiting period while their meal is prepared. The open floor plan is reassuring, and promotes trustworthiness with customers: you can see your meal being prepared, and know that their kitchen is clean and safe.
While waiting for food, customers have a few options to occupy their time: there is a television, artwork, and a gender-neutral restroom.
Customer Journey: Phase 6 - Value
To extract value from the transaction, customers must receive and consume their food. I think this was worth the wait.