Back to Home

JobForge AI

AI/MLFullstack

An AI career tool to automate tailored job applications, resume/cover letter writing, job search etc. I built this because applying job is too annoying.

JobForge AI - Project Planning Document

Project Goal

Build an AI-powered job application assistant that automates the job search and application process by generating tailored resumes and cover letters, searching for relevant positions, and streamlining the application workflow. This project serves as a hands-on learning experience for transitioning from software development to AI engineering.

Project Summary

JobForge AI is an intelligent agent tool designed to help job seekers efficiently manage their job search process. Users provide their professional information, preferences, and career goals, and the AI agent handles the heavy lifting: searching for suitable positions, analyzing job-candidate fit, and generating customized application materials. The system leverages LLMs for document generation and prompt engineering, combines it with web scraping/APIs for job discovery, and provides an intuitive interface for managing the entire application pipeline.

Target User: Job seekers (especially students and early-career professionals) who want to apply to multiple positions efficiently while maintaining quality and personalization.

Core Features (MVP)

1. Data Entry & Management

  • Resume Upload & Scan: Parse existing resumes (PDF/DOCX) to extract structured information
  • User Information Form: Web form for entering/editing professional details:
    • Personal information (name, contact, location)
    • Work experience (company, role, duration, responsibilities)
    • Education (institution, degree, graduation date)
    • Projects (title, description, technologies, links)
    • Skills (technical, soft skills, certifications)
    • Additional sections (awards, publications, volunteer work)
  • Data Storage: PostgreSQL database to persist user profiles

2. User Job Preferences

  • Job Search Criteria Form:
    • Desired job titles/roles
    • Preferred locations (remote, hybrid, on-site, specific cities)
    • Employment type (full-time, part-time, contract, internship)
    • Salary expectations (optional)
    • Start date availability
    • Industry preferences
    • Company size preferences
  • Additional Information Field: Free-text area for specific requirements or preferences

3. Job Search & Analysis

  • Job Discovery: Integration with Arbeitnow API for fetching job listings
  • Filtering System:
    • Filter by position title, category, location, employment type
    • Keyword matching for required skills
  • Job Assessment:
    • Skills match percentage (compare user skills vs. job requirements)
    • Difficulty level estimation (entry/mid/senior alignment)
    • Salary range extraction and comparison
    • Application deadline tracking
  • Results Dashboard: Display filtered and ranked job opportunities

4. AI-Powered Document Generation

  • Tailored Resume Generation:
    • Analyze job description to identify key requirements
    • Customize resume sections to highlight relevant experience
    • Emphasize matching skills and accomplishments
    • Adjust technical terminology to match job posting
  • Custom Cover Letter Generation:
    • Generate personalized cover letters addressing specific job requirements
    • Incorporate company research and role alignment
    • Maintain professional tone while showing genuine interest
  • User Guidelines Support: Allow users to provide specific instructions for document generation
  • Multiple Format Export: PDF and DOCX output formats
  • Version Control: Save multiple versions for different applications

Future Features

Phase 2 - Enhancement Features

  • [ ] AI Voice Chat for Data Entry: Natural conversation interface to collect user information through voice interaction
  • [ ] Automated Job Application: One-click application system that fills out online application forms
    • Intelligent form field detection and mapping
    • User notification for missing required information
    • Application confirmation and tracking
  • [ ] Browser Extension: Chrome/Firefox extension for seamless job board interaction

Phase 3 - Advanced Features

  • [ ] Application Tracking Dashboard:
    • Visual timeline of all applications
    • Status tracking (applied, interviewing, rejected, offer)
    • Notes and follow-up reminders
    • Analytics on application success rates
  • [ ] Interview Preparation Assistant:
    • Generate likely interview questions based on job description
    • Company research summaries
    • STAR method response templates
    • Mock interview practice
  • [ ] Skill Gap Analysis:
    • Compare user skills against job market demands
    • Identify missing skills for target roles
    • Suggest learning resources and projects
    • Track skill development progress
  • [ ] Multi-Version Resume Management:
    • Store multiple resume templates (technical, managerial, industry-specific)
    • A/B testing to track which versions perform better
    • Template recommendations based on job type
  • [ ] Email Draft Generator:
    • Follow-up email templates
    • Thank you notes after interviews
    • Networking message drafts
    • Salary negotiation email templates
  • [ ] Salary Insights & Benchmarking:
    • Aggregate salary data from multiple sources
    • Compensation comparison by role, location, experience
    • Market rate recommendations
  • [ ] Application Analytics Dashboard:
    • Success rate metrics by job type, company size, location
    • Pattern identification in rejections
    • Time-to-response analysis
    • ROI tracking for application efforts

Phase 4 - Integration

  • [ ] Portfolio Website Integration: Seamless connection with portfolio generation tool
    • Auto-sync professional information
    • Generate portfolio links for applications
    • Embed JobForge AI functionality in portfolio sites

Learning Outcomes

By completing this project, I will gain hands-on experience with:

AI Engineering Skills

  1. LLM Integration & API Usage: Working with Azure OpenAI Service (GPT-4o-mini)
  2. Prompt Engineering: Crafting effective prompts for resume and cover letter generation
  3. RAG (Retrieval-Augmented Generation): Combining user data with job descriptions for context-aware generation
  4. Structured Output Generation: Ensuring LLM outputs match required formats
  5. Cost Optimization: Managing API costs and token usage efficiently

Software Engineering Skills

  1. Full-Stack Development: Building with Next.js frontend and FastAPI backend
  2. Database Design: PostgreSQL schema design for complex user data
  3. API Integration: Working with third-party APIs (Arbeitnow, Azure OpenAI)
  4. Web Scraping: Ethical data extraction techniques (backup to API)
  5. Document Processing: PDF/DOCX parsing and generation
  6. State Management: Handling complex application state in React
  7. Asynchronous Processing: Managing long-running AI tasks

Product & System Design

  1. User Experience Design: Creating intuitive workflows for complex processes
  2. System Architecture: Designing scalable, maintainable application structure
  3. Data Privacy: Handling sensitive user information securely
  4. Error Handling: Graceful degradation and user-friendly error messages

Tech Stack

Frontend

  • Framework: Next.js 14 (App Router)
  • Language: TypeScript
  • Styling: Tailwind CSS
  • State Management: React Context / Zustand
  • Forms: React Hook Form + Zod validation
  • UI Components: Shadcn/ui (optional, for consistent design)

Backend

  • Framework: FastAPI (Python 3.10+)
  • ORM: SQLAlchemy
  • Database: PostgreSQL 15+
  • Document Processing:
    • PyPDF2 / pdfplumber (PDF parsing)
    • python-docx (DOCX parsing/generation)
    • ReportLab / WeasyPrint (PDF generation)
  • Web Scraping:
    • Requests (API calls)
    • BeautifulSoup4 (backup scraping if needed)

AI/ML

  • LLM Provider: Azure OpenAI Service
  • Model: GPT-4o-mini
  • SDK: OpenAI Python SDK (compatible with Azure)
  • Prompt Management: LangChain (optional, for complex workflows)

Infrastructure & Deployment

  • Version Control: Git + GitHub
  • Frontend Hosting: Vercel
  • Backend Hosting: Railway / Render (free tier)
  • Database Hosting: Railway PostgreSQL / Supabase
  • Environment Management: python-dotenv, .env files

Development Tools

  • API Testing: Postman / Thunder Client
  • Database Management: pgAdmin / DBeaver
  • Code Quality: ESLint, Prettier, Black (Python formatter)
  • Type Checking: TypeScript, mypy (Python)

Project Timeline & Todos

Week 1-2: Project Setup & Foundation

Goal: Set up development environment and basic architecture

  • Initialize Next.js project with TypeScript and Tailwind CSS
  • Set up FastAPI project structure with virtual environment
  • Configure PostgreSQL database (local + hosted)
  • Set up Azure OpenAI Service account and get API keys
  • Create GitHub repository with proper .gitignore
  • Set up development environment variables
  • Design database schema (users, resumes, job_listings, applications, generated_documents)
  • Create initial SQLAlchemy models
  • Set up CORS and API communication between frontend and backend
  • Create basic project documentation (README.md, setup instructions)

Deliverable: Working development environment with database connectivity

Week 3-4: Data Entry System

Goal: Build user profile management system

  • Create database migration for user tables
  • Build FastAPI endpoints for user CRUD operations
  • Implement resume upload functionality (backend)
  • Build PDF/DOCX parser to extract text and structure
  • Create user information form (Next.js frontend)
    • Personal information section
    • Work experience (dynamic add/remove)
    • Education section
    • Projects section
    • Skills section
  • Implement form validation with Zod
  • Connect frontend forms to backend API
  • Build user profile dashboard page
  • Test data persistence and retrieval

Deliverable: Functional user profile creation and management system

Week 5-6: Job Search Integration

Goal: Implement job discovery and filtering

  • Sign up for Arbeitnow API and test endpoints
  • Build FastAPI service for job search
  • Implement job fetching from Arbeitnow API
  • Create job_listings table and models
  • Build filtering logic (location, type, keywords)
  • Implement skill matching algorithm
  • Create job search interface (Next.js)
    • Search criteria form
    • Filters and sorting options
  • Build job listing cards/table view
  • Implement job details modal/page
  • Add job bookmarking functionality
  • Create job assessment scoring system
    • Skills match percentage
    • Difficulty level indicator
    • Salary alignment (if available)

Deliverable: Working job search and filtering system with assessment scores

Week 7-9: AI Document Generation

Goal: Implement core AI-powered resume and cover letter generation

  • Set up Azure OpenAI SDK in FastAPI
  • Design prompt templates for resume generation
  • Design prompt templates for cover letter generation
  • Implement RAG pattern: combine user data + job description
  • Build document generation endpoints
  • Create prompt engineering system with user guidelines support
  • Implement token usage tracking and cost monitoring
  • Build document preview interface (Next.js)
  • Add PDF export functionality (WeasyPrint/ReportLab)
  • Add DOCX export functionality (python-docx)
  • Implement document versioning and storage
  • Create document editing interface for post-generation tweaks
  • Build A/B testing for different prompt strategies
  • Add document download and share functionality

Deliverable: End-to-end document generation pipeline with export capabilities

Week 10-11: Integration & Polish

Goal: Connect all features and improve user experience

  • Build unified application workflow:
    • Profile setup → Job search → Document generation → Application tracking
  • Create application tracking table/system
    • Save which jobs user applied to
    • Track application status
    • Store generated documents per application
  • Implement dashboard homepage showing:
    • Application statistics
    • Recent job matches
    • Pending applications
  • Add loading states and progress indicators for AI operations
  • Implement error handling and user-friendly error messages
  • Add success notifications and confirmations
  • Optimize API response times
  • Implement basic analytics (jobs viewed, applications created)
  • Add user settings page (preferences, API usage tracking)
  • Create onboarding flow for new users

Deliverable: Polished MVP with smooth user experience

Week 12: Testing & Deployment

Goal: Test thoroughly and deploy to production

  • Write unit tests for critical backend functions
  • Test all API endpoints with various inputs
  • Perform end-to-end user flow testing
  • Test with real job listings and user data
  • Fix identified bugs and edge cases
  • Optimize database queries (add indexes)
  • Set up production environment variables
  • Deploy backend to Railway/Render
  • Deploy frontend to Vercel
  • Set up production PostgreSQL database
  • Configure domain (optional)
  • Test production deployment thoroughly
  • Create user documentation and help guides
  • Write comprehensive README with setup instructions

Deliverable: Deployed, functional MVP available for use

Week 13+: Iteration & Future Features

Goal: Gather feedback and plan next phases

  • Use the application for personal job search
  • Gather feedback from peers/friends
  • Document lessons learned
  • Prioritize Phase 2 features based on usage
  • Plan architecture for automated application feature
  • Research browser extension development
  • Begin portfolio website integration planning

Success Metrics

  • Functional: Successfully generate tailored resume and cover letter for real job posting
  • Performance: Document generation completes in <10 seconds
  • Cost: Stay within Azure student credit limits ($100/year)
  • Learning: Complete at least 3 real job applications using the tool
  • Code Quality: Maintain clean, documented, testable codebase

Resources & References

Documentation

Learning Resources

  • Anthropic's Prompt Engineering Guide
  • OpenAI Prompt Engineering Best Practices
  • FastAPI + Next.js Integration Patterns
  • RAG Implementation Tutorials

Last Updated: January 2026
Version: 1.0
Status: Planning Phase