ResuCraft

• 

• 

• 

• 

• 

• 

✨ Inspiration

Applying to jobs as a student often means sending out dozens—or even hundreds—of applications. But to truly stand out, each resume needs to be customized to the job. That kind of resume tailoring is time-consuming, especially when you're juggling coursework, deadlines, and pressure.

Worse, many students don’t know where to start. Writing a strong resume can be overwhelming if you’ve never been taught how. And even if your resume is good, you're still left guessing what questions you’ll face in the interview.

We created ResuCraft to fix all of that. By using AI to generate personalized, job-specific resumes and simulate realistic interview calls, we give students the tools to prepare smarter and apply faster — with confidence.

💡 What it does

ResuCraft is an AI-powered platform that helps students instantly generate customized resumes and simulate real interview experiences — all tailored to the specific job they're applying for.

With our Chrome extension, students can scrape job listings directly from sites like Indeed with a single click. The job data is automatically sent to the ResuCraft dashboard, where users can view saved listings and select the ones they want to apply to.

From there, ResuCraft uses the latest Gemini model to generate a unique resume for each job — intelligently selecting and highlighting only the most relevant experiences from the student’s profile. No more copy-pasting or guessing what to include.

But we don’t stop at resumes. ResuCraft also simulates live AI phone interviews, where our system does the research for you — analyzing the job description and role requirements to ask the most accurate, job-specific questions. It’s like having a personalized recruiter and interview coach, powered by cutting-edge AI.

🛠 How we built it

Frontend: HTML, CSS, and JavaScript with a responsive, student-friendly dashboard UI. We placed strong emphasis on UI/UX design to make the experience intuitive, fast, and easy for students navigating job applications and resume generation.

Backend: PHP with MongoDB for managing user accounts, job listings, resumes, and AI-generated interview data — optimized for flexibility and speed.

AI Integration: Gemini and GPT-4.1 were used to generate deeply personalized resumes and craft role-specific interview questions based on real job descriptions.

Calling System: Vapi.ai was integrated to simulate real-time AI phone interviews, using LLM-driven voice interactions tailored to each job.

Chrome Extension: A lightweight browser extension that auto-scrapes job descriptions from platforms like Indeed and sends them directly to the dashboard — eliminating manual copy-paste and streamlining the entire application process.

🧩 MLH Sponsor Challenge Participation

We proudly integrated all MLH sponsor technologies in ResuCraft, aligning with their respective challenges:

🌐 MongoDB Atlas Challenge

We used MongoDB as the backbone of our backend, managing everything from user accounts and saved resumes to job listings and AI-generated interview data.

What made MongoDB ideal for ResuCraft was its flexible document schema, which allowed us to store highly variable user data — including unique combinations of education, experiences, certifications, and projects — without needing a rigid or uniform structure. This flexibility was critical, as no two users have the same background, and the data needed to adapt dynamically to each resume and interview flow.

Used for: storing resumes, job data, user profiles, and interview content generated by LLMs

Why MongoDB: schema-less design, scalability, and fast query performance for a responsive, real-time dashboard experience

*🤖 Gemini API Challenge *

We used the Gemini API to generate highly tailored, structured resume content and analyze job descriptions. Gemini helped us extract the most relevant context from job postings, summarize key responsibilities, and guide GPT-4.1 in building focused interview questions.

Used for: summarizing job descriptions and generating custom resume bullet points

Why Gemini: fast, structured output that blends well with prompt-chaining and multi-LLM workflows

By combining these technologies, we were able to create a seamless and intelligent experience that saves students real time — and improves their odds at landing interviews.

🚧 Challenges we ran into Managing multiple LLM outputs for consistency, tone, and accuracy across resume content and interview questions

Simulating realistic interview flows via voice calls — including natural pauses, timing, and dynamic question adaptation

Personalizing resumes without sacrificing structure or professionalism — balancing creativity with clarity

Bypassing job listing restrictions — platforms like Indeed heavily combat scraping bots, so we built a Chrome extension to capture job data client-side and send it directly to our backend for seamless, reliable input

🏆 Accomplishments that we're proud of

Full working MVP with resume builder, job dashboard, and live call simulation

Clean UI for students to navigate everything in one place

Real-time AI call system that feels human and job-specific

Strong feedback from test users: “this saved me hours”

📚 What we learned

How to combine multiple LLMs for complementary strengths

Effective ways to map job descriptions to AI prompts

How small UX changes greatly improve confidence and user engagement

The importance of context-aware AI conversations over generic chatbot flows

🚀 What’s next for ResuCraft

Resume scoring + feedback powered by AI

Voice response analysis for interview calls

API to integrate with career centers and university job portals

Built With

• css
• gemini
• gpt-4.1
• html
• javascript
• php
• restapi