This is a submission for the AI Agents Challenge powered by n8n and Bright Data
What I Built
Job Hunter AI is a revolutionary job search automation workflow that combines the power of Amazon Bedrock Chat Model with n8n AI Agent Node and Amazon Nova Premier - AWS's latest multimodal AI model. This cutting-edge system automatically:
๐ Scrapes job listings from LinkedIn using BrightData's verified nodes
๐ง Extracts recruiter contacts using Nova Premier's advanced reasoning capabilities
๐ Generates personalized cover letters with Amazon Bedrock agent framework
๐ Organizes everything in a comprehensive Google Sheets dashboard
๐ฏ Provides actionable insights with multimodal AI analysis
Revolutionary Features:
- Next-gen AI: Amazon Nova Premier (us.amazon.nova-premier-v1:0) via Amazon Bedrock Chat Model
- Advanced reasoning: Superior contact extraction and pattern recognition
- Tool capabilities: Native function calling through n8n AI Agent Node framework
- Multimodal processing: Handles text, images, and structured data
- Modern architecture: Clean Amazon Bedrock integration with n8n AI Agent nodes
n8n Workflow
GitHub Gist: Job Hunter AI Workflow JSON
The workflow is accessible through n8n's interface and can be triggered manually or scheduled to run automatically. Here's how it works:
- Input your job search criteria (title, location, experience level)
- Automated scraping begins on LinkedIn
- AI analysis extracts key details including recruiter contacts
- Personalized content generation for each job opportunity
- Google Sheets export with all data organized and ready for action
Actual Google Sheets Exported Data:
โ ๏ธ Security Note: All sensitive information has been removed from the workflow file:
- Google Sheet ID replaced with
YOUR_GOOGLE_SHEET_ID
- BrightData credential ID replaced with
YOUR_BRIGHTDATA_CREDENTIAL_ID
- AWS credential ID replaced with
YOUR_AWS_CREDENTIAL_ID
- Google Sheets credential ID replaced with
YOUR_GOOGLE_SHEETS_CREDENTIAL_ID
โ ๏ธ Security Best Practice: Always use temporary AWS credentials with least privilege permissions when testing your n8n Agent with AWS services. Never use root credentials or long-term access keys in development environments. Users must configure their own credentials and Sheet ID before importing the workflow.
Technical Implementation
System Architecture
The workflow follows a sophisticated data pipeline:
Input Layer โ Scraping Layer โ AI Processing โ Output Layer
{
"name": "Job Hunter AI - LinkedIn Automation",
"nodes": [
// Complete workflow with 15+ nodes including:
// - Manual Trigger
// - BrightData scrapers for LinkedIn
// - AI Agent for data extraction
// - Amazon Nova Premier for cover letter generation
// - Google Sheets integration
// - Data normalization and structuring
]
}
n8n AI Agent Configuration
System Instructions:
You are an expert recruiter and data extraction specialist. Analyze job postings and extract structured information with high precision. Always return valid JSON only.
Prompt:
Analyze this job posting and extract detailed information:
Job Title: {{ $json.jobTitle }}
Company: {{ $json.companyName }}
Location: {{ $json.jobLocation }}
Description: {{ $json.description }}
Salary: {{ $json.salary }}
Source: {{ $json.sourcePlatform }}
URL: {{ $json.jobUrl }}
Extract and return ONLY valid JSON with these exact fields:
{
"companyName": "exact company name",
"jobTitle": "exact job title",
"description": "cleaned job description",
"salaryRange": "salary information or 'Not specified'",
"location": "job location",
"workType": "Remote/Hybrid/Onsite/Unknown",
"applicationDeadline": "deadline or 'Not specified'",
"recruiterEmail": "email if found or 'Not found'",
"hiringManagerEmail": "email if found or 'Not found'",
"recruiterLinkedIn": "LinkedIn profile if found or 'Not found'",
"benefits": ["benefit1", "benefit2"],
"keySkills": ["skill1", "skill2"],
"experienceLevel": "Entry/Mid/Senior/Executive",
"jobUrl": "{{ $json.jobUrl }}",
"sourcePlatform": "{{ $json.sourcePlatform }}",
"extractedAt": "{{ new Date().toISOString() }}"
}
Model Choice: Amazon Nova Premier
- Reasoning: Superior performance in structured data extraction and natural language generation
- Context window: Handles large job descriptions effectively
- Multilingual support: Works with international job postings
Memory: Stateless execution with job criteria persistence across workflow steps
Tools Used:
- BrightData Community Node: Professional web scraping via n8n-nodes-brightdata
- Amazon Nova Premier: Latest multimodal AI via Amazon Bedrock integration
- Google Sheets API: Data storage and organization
- n8n Code Nodes: Data transformation and normalization
Bright Data Verified Node
The BrightData Verified Node is the backbone of this workflow, providing:
Reliable Data Collection:
- LinkedIn Jobs Scraper: Extracts job listings, company details, and posting metadata
- Anti-detection technology: Ensures consistent data collection without blocks
- Structured data output: Clean, normalized job data ready for AI processing
Bright Data Account Proxy Infrastructure Configuration
Bright Data Verified Node Implementation Details:
// LinkedIn scraping configuration
{
"parameters": {
"zone": {
"__rl": true,
"value": "mcp_unlocker",
"mode": "list",
"cachedResultName": "mcp_unlocker"
},
"country": {
"__rl": true,
"mode": "list",
"value": "us"
},
"url": "=https://www.linkedin.com/jobs/search/?keywords={{ encodeURIComponent($json.jobTitle) }}&location={{ encodeURIComponent($json.location) }}&f_TPR=r86400",
"format": "json",
"requestOptions": {}
},
"id": "97295b11-8c10-48c2-9215-9fb413e1c253",
"name": "LinkedIn Web Unlocker",
"type": "@brightdata/n8n-nodes-brightdata.brightData",
"position": [
-1952,
-272
],
"typeVersion": 1,
"credentials": {
"brightdataApi": {
"id": "YOUR_BRIGHTDATA_CREDENTIAL_ID",
"name": "BrightData account"
}
}
}
Data Quality Assurance:
- Duplicate detection: Prevents duplicate job entries across platforms
- Data validation: Ensures all required fields are populated
- Error handling: Graceful fallbacks when scraping encounters issues
- Rate limiting: Respects platform guidelines while maximizing data collection
Journey
Development Process
Phase 1: Research & Planning
- Analyzed existing job search tools and identified pain points
- Researched BrightData capabilities and n8n integration patterns
- Designed workflow architecture and data flow
Phase 2: Core Implementation
- Built initial scraping workflow for LinkedIn
- Integrated BrightData Verified Node with proper error handling
- Developed data normalization and cleaning processes
Phase 3: Next-Gen AI Integration
- Implemented Amazon Bedrock with Amazon Nova Premier model
- Integrated native n8n Bedrock nodes for superior reasoning capabilities
- Created sophisticated prompts leveraging multimodal AI features
- Built advanced cover letter generation with structured output
Phase 4: Polish & Testing
- Added Google Sheets integration for comprehensive tracking
- Implemented skills gap analysis and interview preparation features
- Extensive testing across different job types and locations
Challenges Overcome
1. Data Consistency Optimization
- Challenge: LinkedIn job data required careful parsing and normalization
- Solution: Built robust normalization layer that handles various data formats
- Learning: Always plan for data variability in web scraping projects
2. Recruiter Contact Extraction
- Challenge: Contact information is often hidden or requires inference
- Solution: Used AI to analyze job descriptions, company pages, and posting patterns
- Learning: LLMs excel at pattern recognition in unstructured text
3. Next-Gen AI Architecture
- Challenge: Traditional AI integrations lack modern capabilities
- Solution: Used Amazon Bedrock Chat Model with n8n AI Agent Node for cutting-edge performance
- Learning: Modern AI frameworks provide superior developer experience and results
4. Cover Letter Personalization
- Challenge: Generic templates don't stand out to recruiters
- Solution: AI analyzes both job requirements and user resume for perfect matching
- Learning: Context-aware AI generation significantly improves output quality
What I Learned
Technical Insights:
- BrightData Community Node provides reliable, scalable web scraping
- Amazon Nova Premier delivers state-of-the-art reasoning capabilities
- Amazon Bedrock integration enables native n8n AI workflows
- Multimodal AI opens new possibilities for job data analysis
- Modern architecture improves maintainability and performance
Product Insights:
- Job seekers need more than just job listings - they need actionable intelligence
- Personalization at scale is possible with the right AI tools
- Contact information is often the missing piece in job applications
- Comprehensive tracking helps job seekers stay organized and strategic
Business Insights:
- Time-to-application is critical in competitive job markets
- Quality over quantity - better to apply to fewer jobs with personalized materials
- Data-driven job searching provides significant competitive advantages
- Automation frees up time for networking and skill development
Future Enhancements
Planned Features:
- Salary negotiation insights based on market data
- Company culture analysis from employee reviews
- Application tracking with follow-up reminders
- Interview scheduling automation
- Skills recommendation engine based on job market trends
Technical Improvements:
- Real-time notifications for new matching jobs
- Mobile app integration for on-the-go job management
- Advanced filtering with ML-based job matching
- Integration with ATS systems for direct application submission
Built with โค๏ธ using n8n and BrightData
Transforming job search from a tedious process into an intelligent, automated system that helps candidates find their dream roles faster and more effectively.
Top comments (0)