from jobx import scrape_jobs # Search for Python developer jobs jobs = scrape_jobs( search_term="python developer", location="New York, NY", results_wanted=50 ) print(f"Found {len(jobs)} jobs") print(jobs[["title", "company", "location", "confidence_score"]].head()) # Save to different formats jobs.to_csv("jobs.csv", index=False) jobs.to_parquet("jobs.parquet", index=False)

from jobx import scrape_jobs jobs = scrape_jobs( search_term="python developer", location="New York", results_wanted=100 ) # Jobs are automatically sorted by confidence score (highest first) print(jobs[['title', 'company', 'confidence_score']].head(10)) # Filter for highly relevant jobs (70%+ confidence) relevant_jobs = jobs[jobs['confidence_score'] >= 0.7] print(f"Found {len(relevant_jobs)} highly relevant jobs out of {len(jobs)} total") # Analyze confidence distribution print(f"Average confidence: {jobs['confidence_score'].mean():.2%}") print(f"Jobs with 80%+ confidence: {len(jobs[jobs['confidence_score'] >= 0.8])}")

from jobx import scrape_jobs # Track SERP positions and identify your company's postings jobs = scrape_jobs( search_term="machine learning engineer", location="Boston", track_serp=True, my_company_names=["Acme Corp", "Acme Inc."], results_wanted=100 ) # Analyze SERP visibility print(f"Average position: {jobs['serp_absolute_rank'].mean():.1f}") print(f"Page 1 listings: {len(jobs[jobs['serp_page_index'] == 0])}") print(f"Sponsored posts: {jobs['serp_is_sponsored'].sum()}") # Compare your company vs competitors my_company_jobs = jobs[jobs['is_my_company'] == True] competitor_jobs = jobs[jobs['is_my_company'] == False] print(f"Your company: {len(my_company_jobs)} postings") print(f"Average rank: {my_company_jobs['serp_absolute_rank'].mean():.1f}") print(f"Competitors: {len(competitor_jobs)} postings")

from jobx import scrape_jobs from jobx.model import Site, JobType jobs = scrape_jobs( site_name=[Site.LINKEDIN, Site.INDEED], search_term="software engineer", location="San Francisco, CA", distance=50, job_type=JobType.FULL_TIME, is_remote=True, easy_apply=True, # LinkedIn only results_wanted=200, enforce_annual_salary=True, linkedin_fetch_description=True, verbose=2 )

# Filter high-paying remote positions high_paying_remote = jobs[ (jobs['is_remote'] == True) & (jobs['min_amount'] >= 120000) & (jobs['currency'] == 'USD') & (jobs['interval'] == 'yearly') ] # Group by company and analyze company_stats = high_paying_remote.groupby('company').agg({ 'min_amount': ['count', 'mean'], 'max_amount': 'mean', 'title': lambda x: ', '.join(x.unique()[:3]) }).round(0) print(company_stats) # Save filtered results as Parquet for efficient storage high_paying_remote.to_parquet("high_paying_remote_jobs.parquet", compression='snappy', index=False)

# Core fields job_columns = [ 'title', 'company', 'location', 'job_url', 'description', 'date_posted', 'is_remote', 'job_type', 'site', 'min_amount', 'max_amount', 'currency', 'interval', 'salary_source', 'emails', 'easy_apply', 'confidence_score' ] # Location details location_fields = ['city', 'state', 'country'] # Compensation analysis salary_analysis = jobs.groupby('site').agg({ 'min_amount': ['count', 'mean', 'median'], 'max_amount': ['mean', 'median'], 'is_remote': 'sum' })