I gave Claude one app and asked it to act like four different senior developers — here’s what happened
I asked Claude to build, debug, redesign and optimize the same app as four senior developers. Here’s what each role fixed — and missed.
David Max, an AI educator, shared a collection of prompts on the internet that claimed to transform Claude into various senior developers, such as a debugging engineer, frontend engineer, and performance engineer. Intrigued yet skeptical, the reporter decided to test these prompts and observe the results.
First, the reporter asked Claude to create a household expense tracker. The prompt instructed Claude to act like a senior full-stack engineer and develop a polished startup MVP. After outlining the architecture, data structure, and user flow, Claude delivered a surprisingly polished React app with features like expense addition and deletion, categories, filtering, search, sorting, and persistence. The app boasted stat cards, a clean expense form, colored bars for spending breakdown, and responsiveness.
Next, the reporter requested Claude to act as a senior debugging engineer. The goal was to examine the app for functional bugs, invalid input handling, data-loss risks, incorrect calculations, persistence problems, and edge cases. Claude discovered several issues, including improperly structured inputs, missing form submission handling, date validation errors, incorrect calculations, hidden storage failures, and the absence of automated tests.
The debugging engineer provided a detailed report of the issues, their causes, severity levels, and recommended fixes. The fixes were then implemented without redesigning the interface or making broader architectural changes.
Finally, the reporter asked Claude to assume the role of a senior frontend engineer specializing in responsive design and accessibility. The app was reviewed from the perspective of a user on a phone, using a keyboard, or with assistive technology. Claude found that the app lacked visible focus indicators for keyboard navigation, improper labeling connections, a screen reader-unfriendly search field, and insufficient color contrast.
After making improvements like adding focus indicators, correcting label connections, providing screen reader descriptions, and increasing text contrast, Claude ensured that the app was usable and accessible across various devices and user preferences.
Written by urgent.news from Tom's Guide's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.