DEV Community

Cover image for How I Built LaunchTrack: An AI Career Coach with Next.js and Gemini API
Vipin Yadav
Vipin Yadav

Posted on • Originally published at devxvipin.me

How I Built LaunchTrack: An AI Career Coach with Next.js and Gemini API

Hey everyone! πŸ‘‹

I'm excited to share my latest project - **LaunchTrack**, an AI-powered career coaching platform. In this post, I'll walk you through how I built it and the technologies I used.

## πŸš€ What is LaunchTrack?

LaunchTrack is an AI career coach that helps job seekers with:

- Personalized job search support
- Interview preparation
- Resume analysis
- Career advice powered by AI

**Live Demo:** [launchtrack.vercel.app](https://launchtrack.vercel.app)
**GitHub:** [github.com/vipinyadav01/aicareercoach](https://github.com/vipinyadav01/aicareercoach)

## πŸ› οΈ Tech Stack

- **Frontend:** Next.js 14 (App Router)
- **AI:** Google Gemini API
- **Database:** Prisma + PostgreSQL
- **Authentication:** Clerk
- **Styling:** Tailwind CSS + Shadcn UI
- **State Management:** Zustand
- **Background Jobs:** Inngest

## πŸ’‘ Key Features

### 1. AI-Powered Career Advice

Using Google Gemini API, users get personalized career guidance based on their profile and goals.

### 2. Resume Analysis

Upload your resume and get instant AI feedback on how to improve it.

### 3. Interview Preparation

Practice with AI-generated interview questions tailored to your target role.

### 4. PWA Support

Works offline and can be installed as a mobile app!

## πŸ”§ Architecture Overview
Enter fullscreen mode Exit fullscreen mode

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚ Next.js App β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚ Pages β”‚ API Routes β”‚ Server Actions β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚ Prisma ORM β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚ PostgreSQL β”‚ Gemini API β”‚ Inngest β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜


## 🎯 Challenges & Solutions

### Challenge 1: Rate Limiting Gemini API
**Solution:** Implemented request queuing with Inngest background jobs.

### Challenge 2: Streaming AI Responses
**Solution:** Used Next.js Edge Runtime with streaming responses.

### Challenge 3: PWA Offline Support
**Solution:** Configured next-pwa with custom service worker caching.

## πŸ“Š Results

- ⚑ **Lighthouse Score:** 95+
- πŸ“± **PWA:** Installable on all devices
- πŸ”’ **Security:** A+ rating

## πŸ”— Links

- **Portfolio:** [devxvipin.me](https://devxvipin.me)
- **GitHub:** [github.com/vipinyadav01](https://github.com/vipinyadav01)
- **LinkedIn:** [linkedin.com/in/vipinyadav01](https://linkedin.com/in/vipinyadav01)

---

Thanks for reading! Let me know if you have any questions in the comments.

Follow me for more web development content! πŸš€

#nextjs #ai #webdev #javascript
Enter fullscreen mode Exit fullscreen mode

Top comments (0)