Inspiration
Creating a resume from scratch is a notorious bottleneck in the job hunt, often resulting in "formatting nightmares" and inconsistent messaging. I wanted to bridge the gap between a user’s professional digital footprint and a polished, ATS-friendly document. By leveraging AI to handle the heavy lifting of content optimization, I aimed to turn a multi-hour chore into a 30-second workflow.
What it does
MakeResume AI is a browser extension that transforms web-based professional profiles into job-ready documents. It scrapes data directly from the DOM, utilizes Gemini AI to optimize descriptions for specific job keywords, and dynamically generates or edits .docx files in real-time.
How we built it
Architecture: Built on Chrome Extension Manifest V3 for modern security and performance standards.
AI Engine: Integrated the Gemini API to handle complex text summarization and professional "tone-shifting."
Document Logic: Implemented Docx.js to programmatically construct OpenXML documents (Word) directly in the browser memory.
Extraction: Developed custom JavaScript selectors to parse and structure unstructured data from various professional networking sites.
Challenges we ran into
The primary hurdle was navigating the Browser Sandbox. Since extensions are restricted from direct file-system access, I had to architect a pipeline using Service Workers and Blob URLs. This allowed the extension to generate binary document data in the background and "hand it off" to the browser's download manager without requiring an external server, preserving user privacy.
Accomplishments that we're proud of
I’m most proud of achieving a zero-friction "one-click" workflow. Moving from a raw HTML webpage to a structured, professionally formatted Word document—with AI-enhanced content—in a single click is a significant UX win that completely eliminates the "copy-paste" fatigue of resume building.
What we learned
This project was a deep dive into Asynchronous JavaScript and the nuances of the Manifest V3 lifecycle. I also gained significant experience in Prompt Engineering, specifically learning how to constrain LLM outputs to ensure they remain professional, factual, and correctly formatted for resume standards.
What's next for MakeResume AI
The roadmap includes a Template Engine to allow users to toggle between different aesthetic styles (Creative vs. Executive). I am also working on a "Live Match" Sidebar that calculates a JD Score in real-time, showing users exactly which keywords they need to add to beat the bots.
Log in or sign up for Devpost to join the conversation.