PrepPilot AI 🚀

Turn any resume into a personalized interview prep plan in minutes

PrepPilot AI is a complete interview preparation platform that transforms static resumes into interactive, personalized preparation systems. Upload your resume, paste a job description, and receive instant fit scores, tailored resume bullets, mock interview sessions, and 7-day prep plans.

Inspiration

We built PrepPilot AI because job searching can feel fragmented and overwhelming: one tool for resumes, another for interview questions, and a bunch of tabs open at once. We wanted to create something that feels like a real career companion, helping students and early-career job seekers go from "I applied" to "I’m ready" in one smooth workflow. The goal was to make interview prep feel less intimidating and a lot more actionable.

What it does

PrepPilot AI takes a resume and job description, then generates a fit score, missing skill insights, and tailored resume improvements. From there, it creates likely interview questions, a 7-day prep plan, and a mock interview coach that adapts to the user’s answers in real time. Users can also save sessions, export a polished report, and optionally add prep tasks to their calendar.

How we built it

We built the frontend with Next.js 14, TypeScript, Tailwind CSS, shadcn/ui, and Framer Motion to keep the experience fast and polished. The backend uses FastAPI, Prisma, SQLite, and a local OpenAI-compatible mock LLM layer for structured AI workflows, while pdf-parse and mammoth handle resume ingestion from PDF and DOCX files. We also used Zustand, React Hook Form, Zod, Recharts, Sonner, and next-themes to make the app feel responsive, validated, and production-like.

Challenges

One of the biggest challenges was turning messy resume and job description inputs into reliable structured data without breaking the user flow. Another tricky part was designing the mock interview coach so it felt conversational and adaptive instead of just like a static question generator. We also spent time making sure the app stayed visually clean while supporting a lot of functionality in a single workflow.

Accomplishments

We’re proud that PrepPilot AI feels cohesive end-to-end, from landing page to onboarding to analysis to interview practice. The resume-to-prep workflow is genuinely useful, and the mock interview experience adds a layer that makes the project feel much more than a simple AI summarizer. We also managed to keep the product polished enough that it feels like something a real user could trust and come back to.

What we learned

We learned a lot about designing AI-assisted products that are actually helpful instead of just impressive on paper. Building a good experience meant thinking carefully about prompt structure, validation, state management, and how to present AI output in a way that users can act on quickly. We also got better at balancing speed of iteration with product quality during a hackathon timeline.

What's next

Next, we’d love to improve the accuracy of the fit scoring with richer parsing and more role-specific evaluation. We also want to add real calendar integrations, stronger session analytics, and more personalized coaching based on interview performance over time. Longer term, we’d like to support multiple career tracks and make PrepPilot AI a full job-search companion.

Built With

• docker
• fastapi
• framer-motion
• lucide-react
• mammoth
• next-themes
• next.js-14
• openai-compatible-local-mock-llm-layer
• pdf-parse
• prisma
• python-3.11
• react-hook-form
• recharts
• shadcn/ui
• sonner
• sqlite
• tailwind-css
• typescript
• zod
• zustand