Skip to content
BranchesTags

Repository files navigation

🚀 PrepTrack AI — Final Commit · 4:10 AM

"Shipped at the edge of exhaustion. Built with intention."

PrepTrack AI is a comprehensive, AI-powered interview and career preparation platform built with React 19, TypeScript, Vite, and the Google Gemini API. It helps students and professionals get job-ready through structured learning paths, live mock interviews, resume analysis, AI mentoring, and more — all in one unified interface.

📸 What This Commit Includes

This is the final stable build (backup for best 17-11) — timestamped April 11, 2026 at 17:11, featuring:

  • ✅ Video subtitles fixed — Interview Simulator subtitle sync corrected
  • ✅ Video player fixed — Embedded video playback issues resolved
  • ✅ All core features fully functional and tested

✨ Features

Feature Description
🎯 Personalized Onboarding Select domains, languages, and target companies on first launch
📊 Dashboard Track progress across all active learning paths with visual indicators
📚 Curriculum View Structured topic-by-topic learning with videos, notes, and practice problems
🎤 Interview Simulator 4-round AI mock interview with TTS voice questions, cheating detection, and a full hiring report
🤖 Mentor AI Personal AI mentor for on-demand guidance, doubt resolution, and roadmap advice
📄 Resume Analyzer Upload a resume and get an AI-powered analysis with skill gap insights
💼 Job Match Match your profile against job descriptions with compatibility scoring
🔭 Tech Scout Discover trending technologies and tools relevant to your domain
📝 Mock Assessment Timed MCQ assessments with instant scoring and feedback
💻 Code Challenge AI-generated coding problems with automated evaluation

🛠️ Tech Stack

  • Frontend: React 19, TypeScript, Tailwind CSS
  • Build Tool: Vite 6
  • AI Engine: Google Gemini 2.5 Flash (@google/genai)
    • Text generation, TTS voice synthesis, behavior analysis
  • Storage: localStorage for user progress and session persistence

📁 Project Structure

├── App.tsx                    # Root router & global state manager
├── index.tsx                  # React entry point
├── types.ts                   # Shared TypeScript interfaces & types
├── components/
│   ├── Landing.tsx            # Auth / login screen
│   ├── Onboarding.tsx         # Domain & preference setup
│   ├── Dashboard.tsx          # Main hub with progress overview
│   ├── CurriculumView.tsx     # Topic-level learning path
│   ├── InterviewSim.tsx       # 4-round mock interview engine (TTS + video)
│   ├── MentorAI.tsx           # Conversational AI mentor
│   ├── ResumeAnalyzer.tsx     # PDF resume upload & analysis
│   ├── JobMatch.tsx           # JD-to-profile matching
│   ├── TechScout.tsx          # Technology discovery tool
│   ├── MockAssessment.tsx     # Timed MCQ quiz engine
│   └── CodeChallenge.tsx      # Coding problem generator & evaluator
├── services/
│   ├── geminiService.ts       # Gemini API calls (TTS, interview questions, reports)
│   ├── behaviorAnalysis.ts    # Cheating / behavior detection logic
│   ├── onDemandService.ts     # On-demand AI generation helpers
│   └── pdfExtractor.ts        # Resume PDF text extraction
├── data/
│   └── curriculum.ts          # Static learning path content for all domains
├── index.html
├── vite.config.ts
└── tsconfig.json

⚙️ Getting Started

Prerequisites

  • Node.js v18+
  • A valid Google Gemini API key

Installation

# 1. Clone / unzip the project
cd preptrack-ai

# 2. Install dependencies
npm install

# 3. Set up your environment
cp .env.local.example .env.local
# Then add your key:
# VITE_API_KEY=your_gemini_api_key_here

# 4. Start the dev server
npm run dev

Build for Production

npm run build
npm run preview

🌐 Supported Domains

PrepTrack supports structured learning paths across 5 domains:

  • DSA — Data Structures & Algorithms (Python, C++, Java)
  • ML — Machine Learning (Python, TensorFlow, PyTorch)
  • Data Science — Data analysis, visualization, SQL
  • Cyber Security — Network security, ethical hacking, cryptography
  • Web Dev — Frontend, Backend, Full Stack

Target companies: Google, Amazon, Core ECE

🔑 Environment Variables

Variable Purpose
API_KEY Google Gemini API key (required for all AI features)

🧠 Interview Simulator — How It Works

  1. Upload your resume (PDF) before starting
  2. The AI reads your resume and tailors questions to your experience
  3. 4 rounds are conducted:
    • Round 1: Fundamentals
    • Round 2: Application & Scenarios (resume-specific)
    • Round 3: Advanced / Problem Solving (company tech stack)
    • Round 4: HR & Communication
  4. Voice questions are spoken aloud via Gemini TTS (Kore voice)
  5. On completion, a full hiring report is generated including scores, skill breakdown, behavior flags, and a hire/no-hire recommendation

📌 Notes

  • User progress is saved to localStorage under preptrack_progress_v3 and preptrack_user_v3
  • Video subtitles and playback have been patched in this final commit
  • The .env.local file is included in .gitignore — never commit your API key

Third Commit — April 12, 2026 · 12:15 AM What's New in This Commit 🚀 AI Engine & Performance Optimization Transitioned to Production-Grade AI: Replaced prototype mock engines with a pluggable architecture in services/geminiService.ts, fully integrating Gemini for richer, context-aware feedback.

Optimized Build Pipeline: Refined vite.config.ts and build scripts for a more lightweight production footprint and faster HMR (Hot Module Replacement) during development.

📈 Smart Job Discovery & Dashboard Integration Contextual Job Matching: Implemented a new recommendation engine that suggests real-world roles based on curriculum completion percentages and skills validated through the AI Interview Simulator.

Enhanced Dashboard UI: Redesigned the activity control center to visualize upcoming milestones and track multi-domain mastery assessments in a single view.

🛠️ Core Infrastructure & Stability Schema Validation: Hardened the data layer in data/curriculum.ts to ensure consistency across the learning paths for DSA, Web Dev, ML, Data Science, and Cyber Security.

Improved State Persistence: Enhanced localStorage synchronization to prevent data loss during multi-step mock tests and long-form interview simulations.

Unified Branding: Applied a cohesive UI/UX theme across all new modules (InterviewSim, MentorAI, and ResumeAnalyzer) using Tailwind CSS for a polished, "hackathon-ready" finish.

📝 Documentation & Project Layout New Project Manifest: Added a comprehensive README.md featuring a quick-start guide, tech stack breakdown, and high-level project architecture.

Refactored Component Tree: Reorganized the components/ directory for better modularity, separating core views (Landing, Onboarding) from interactive widgets.

Second Commit — April 11, 2025 · 9:30 PM

What's New in This Commit

🏗️ Full Rewrite — React 18 → React 19 + Vite + TypeScript

  • Migrated from Create React App to Vite with full TypeScript support
  • Upgraded to React 19 and @google/genai SDK (replaces direct fetch calls to Gemini REST API)
  • New strongly-typed data models in types.ts covering domains, learning paths, topics, interview reports, and user preferences

🎯 Onboarding Flow (New)

  • Multi-step onboarding screen on first login
  • Users select their domains (DSA, ML, Data Science, Cyber Security, Web Dev)
  • Per-domain configuration: language choice, skill level, libraries, focus areas
  • Preferences saved to localStorage and used to personalise the entire dashboard

📚 Curriculum & Learning Paths (New — data/curriculum.ts)

  • Structured, domain-specific learning paths with topics, descriptions, resources, and practice problems
  • Resources include videos, notes, and projects per topic
  • Progress tracked via topic completion toggles, persisted in localStorage

🎙️ AI Interview Simulator — Major Upgrade (InterviewSim.tsx)

  • Now supports resume upload (paste text) to personalise questions
  • 4-round interview simulation targeting Google, Amazon, or Core ECE roles
  • Round breakdown: Fundamentals → Application → Advanced Problem Solving → HR
  • AI-generated questions adapt based on previous answers (follow-up logic)
  • Text-to-speech via Gemini 2.5 Flash TTS — interviewer voice reads each question aloud
  • Comprehensive final report: overall score, grade, per-round scores, skill breakdown (conceptual, problem solving, communication, confidence, practical)
  • Cheating detection flags, resume skill verification, hiring recommendation (Hire / Borderline / Needs Improvement)
  • Score charts (pie + bar) in final report

🤖 MentorAI (New — MentorAI.tsx)

  • Personalised study roadmap generator
  • User inputs domains, level per domain, daily study hours, duration, and weak areas
  • AI generates a structured, week-by-week preparation roadmap
  • Powered by an on-demand session service (onDemandService.ts)

📄 AI Resume Analyzer — Now Fully Live (ResumeAnalyzer.tsx)

  • Paste resume text + specify target job role
  • AI returns a score, missing skills, and actionable improvement tips
  • Parsed structured report with visual score display
  • Previously a "coming soon" stub — now fully implemented

🧪 Mock Test (MockTest.tsx)

  • Retained from v0.1, integrated into new routing and type system

⚡ PrepTrack AI

AI-powered interview preparation platform — built for students and job seekers who want to practice smarter. First Commit — April 11, 2025 · 6:30 PM

What is PrepTrack AI?

PrepTrack AI is a web application that helps candidates prepare for job interviews using Google's Gemini AI. Instead of reading generic tips, users get a live, interactive experience — real questions, real feedback, real scores. Built as a hackathon project. Fast, focused, and functional.

Features (v0.1 — this commit)

✅ Authentication

  • Login and Register screens
  • localStorage-based session management
  • Protected routes (redirects unauthenticated users)

✅ AI Interview

  • Select a job role (Software Engineer, Data Scientist, PM, etc.)
  • AI asks 5 real interview questions via Gemini
  • Submit your answers and get instant written feedback
  • Full session summary at the end

✅ AI Mock Test

  • Choose a topic (JavaScript, Python, DSA, SQL, React, ML, and more)
  • Gemini generates 5 fresh MCQ questions every time
  • Live answer selection with correct/wrong highlighting
  • Final score with per-question breakdown

🚧 AI Resume Analyzer (button present, coming soon)

🚧 AI Job Match (button present, coming soon)

Tech Stack

Layer Tech
Frontend React.js 18 + React Router v6
AI Google Gemini API (gemini-1.5-flash)
Styling Custom CSS (no UI library)
Auth localStorage (demo-grade)
Build Create React App

Getting Started

# 1. Clone the repo
git clone https://github.com/YOUR_USERNAME/preptrack-ai.git
cd preptrack-ai
# 2. Install dependencies
npm install
# 3. Add your Gemini API key
cp .env.example .env
# Open .env and paste your key
# Get one free at: https://aistudio.google.com/app/apikey
# 4. Run locally
npm start

No API key? The app still works — gemini.js has built-in mock responses so every screen is fully demonstrable without a key.

Project Structure

preptrack-ai/
├── public/
│   └── index.html
├── src/
│   ├── App.jsx              ← Router + protected routes
│   ├── index.js
│   ├── index.css            ← Global design tokens
│   ├── pages/
│   │   ├── Login.jsx        ← Auth (login + register)
│   │   ├── Dashboard.jsx    ← Home with all feature cards
│   │   ├── AIInterview.jsx  ← Live AI interview session
│   │   └── MockTest.jsx     ← AI-generated MCQ test
│   └── utils/
│       └── gemini.js        ← Gemini API calls + mock fallback
├── .env.example
├── .gitignore
└── README.md

Roadmap — Upcoming Commits

v0.2 — AI Resume Analyzer

  • Upload resume as PDF
  • Gemini reads and scores it against a target job role
  • Highlights weak sections, missing keywords, and formatting issues
  • Downloadable feedback report

v0.3 — AI Job Match

  • Input your skills and experience level
  • AI matches you to job roles and explains the fit
  • Suggests which skills to build next

v0.4 — User Dashboard & Progress Tracking

  • Interview history saved across sessions
  • Mock test scores plotted over time
  • Streak tracking and weekly goals

v0.5 — Better Auth

  • Firebase or Supabase authentication
  • User profiles with photo and bio
  • Replace localStorage with real sessions

v1.0 — Polish & Deploy

  • Fully responsive mobile layout
  • Dark mode toggle
  • Deployed on Vercel
  • Custom domain

Contributing

This is a hackathon project in active development. PRs and issues welcome after v0.1 stabilises.

⚡ PrepTrack AI

AI-powered interview preparation platform — built for students and job seekers who want to practice smarter.

First Commit — April 11, 2025 · 6:30 PM

What is PrepTrack AI?

PrepTrack AI is a web application that helps candidates prepare for job interviews using Google's Gemini AI. Instead of reading generic tips, users get a live, interactive experience — real questions, real feedback, real scores.

Built as a hackathon project. Fast, focused, and functional.

Features (v0.1 — this commit)

✅ Authentication

  • Login and Register screens
  • localStorage-based session management
  • Protected routes (redirects unauthenticated users)

✅ AI Interview

  • Select a job role (Software Engineer, Data Scientist, PM, etc.)
  • AI asks 5 real interview questions via Gemini
  • Submit your answers and get instant written feedback
  • Full session summary at the end

✅ AI Mock Test

  • Choose a topic (JavaScript, Python, DSA, SQL, React, ML, and more)
  • Gemini generates 5 fresh MCQ questions every time
  • Live answer selection with correct/wrong highlighting
  • Final score with per-question breakdown

🚧 AI Resume Analyzer (button present, coming soon)

🚧 AI Job Match (button present, coming soon)

Tech Stack

Layer Tech
Frontend React.js 18 + React Router v6
AI Google Gemini API (gemini-1.5-flash)
Styling Custom CSS (no UI library)
Auth localStorage (demo-grade)
Build Create React App

Getting Started

# 1. Clone the repo
git clone https://github.com/YOUR_USERNAME/preptrack-ai.git
cd preptrack-ai

# 2. Install dependencies
npm install

# 3. Add your Gemini API key
cp .env.example .env
# Open .env and paste your key
# Get one free at: https://aistudio.google.com/app/apikey

# 4. Run locally
npm start

No API key? The app still works — gemini.js has built-in mock responses so every screen is fully demonstrable without a key.

Project Structure

preptrack-ai/
├── public/
│   └── index.html
├── src/
│   ├── App.jsx              ← Router + protected routes
│   ├── index.js
│   ├── index.css            ← Global design tokens
│   ├── pages/
│   │   ├── Login.jsx        ← Auth (login + register)
│   │   ├── Dashboard.jsx    ← Home with all feature cards
│   │   ├── AIInterview.jsx  ← Live AI interview session
│   │   └── MockTest.jsx     ← AI-generated MCQ test
│   └── utils/
│       └── gemini.js        ← Gemini API calls + mock fallback
├── .env.example
├── .gitignore
└── README.md

Roadmap — Upcoming Commits

v0.2 — AI Resume Analyzer

  • Upload resume as PDF
  • Gemini reads and scores it against a target job role
  • Highlights weak sections, missing keywords, and formatting issues
  • Downloadable feedback report

v0.3 — AI Job Match

  • Input your skills and experience level
  • AI matches you to job roles and explains the fit
  • Suggests which skills to build next

v0.4 — User Dashboard & Progress Tracking

  • Interview history saved across sessions
  • Mock test scores plotted over time
  • Streak tracking and weekly goals

v0.5 — Better Auth

  • Firebase or Supabase authentication
  • User profiles with photo and bio
  • Replace localStorage with real sessions

v1.0 — Polish & Deploy

  • Fully responsive mobile layout
  • Dark mode toggle
  • Deployed on Vercel
  • Custom domain

Contributing

This is a hackathon project in active development. PRs and issues welcome after v0.1 stabilises.

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

Packages

Contributors

Languages