JobSpy

1.1.82 · active · verified Thu Apr 16

JobSpy is a Python library designed for scraping job postings from major job boards including LinkedIn, Indeed, Glassdoor, ZipRecruiter, Google Jobs, Bayt, and Naukri. It aggregates job data into a Pandas DataFrame, supports concurrent scraping, and includes features like proxy support to manage rate limiting. The library is actively maintained, with frequent updates adding new features and improving scraper reliability.

Common errors

```
Received a response code 429
```
cause The job board site has temporarily blocked your IP due to too many requests.

fix Wait some time between scrape requests, try using the `proxies` parameter with a proxy list, or consider `use_creds=True` for authenticated access where applicable.
```
Segmentation fault: 11 on macOS Catalina
```
cause This issue is typically related to the `tls_client` dependency not fully supporting the specific macOS architecture or version.

fix Upgrade to a newer version of macOS if possible, or refer to the `tls_client` repository for known workarounds or fixes.
```
No results when using 'google'
```
cause The `google_search_term` parameter requires a very specific syntax, usually the exact search string that appears in the Google Jobs search box after applying filters in a browser.

fix Perform the desired search on Google Jobs in your browser, then copy and paste the exact query from the Google Jobs search bar into the `google_search_term` parameter.
```
Indeed giving unrelated roles
```
cause Indeed's search engine often includes keywords from job descriptions, not just titles, which can lead to broader, less relevant results.

fix Refine your `search_term` using exact phrase matching (e.g., `"engineering intern"`) and negation (e.g., `-marketing`) to narrow down results.

Warnings

gotcha Default logging verbosity changed. Logs are now suppressed by default, showing only errors.
Fix: Set `verbose=1` (for warnings+errors) or `verbose=2` (for all logs) in the `scrape_jobs` function call to re-enable detailed logging.
gotcha Indeed job sorting changed from date to relevance by default, which may affect expected results.
Fix: If exact date-based sorting is crucial, consider filtering results by `hours_old` or refining `search_term` for more precise matches.
gotcha Job boards aggressively block IP addresses for too many requests, leading to `response code 429` (rate limiting).
Fix: Implement delays between `scrape_jobs` calls, use proxies (via the `proxies` parameter), or enable authenticated scraping where available (e.g., `use_creds=True` for LinkedIn).
gotcha When searching Indeed or Glassdoor, the `country_indeed` parameter is often required for specific countries to yield correct results.
Fix: Always specify `country_indeed='USA'` or the appropriate country code when scraping Indeed or Glassdoor.
gotcha Certain parameters like `hours_old` cannot be combined with other filtering parameters (e.g., `job_type` + `is_remote`) for specific sites like Indeed or LinkedIn.
Fix: Consult the `scrape_jobs` documentation to understand parameter incompatibilities for each job board and choose only one of the conflicting filters.

Install

pip install -U python-jobspy Install latest version

Imports

scrape_jobs
```
from jobspy import scrape_jobs
```

Quickstart

This quickstart example demonstrates how to scrape job postings for 'software engineer' in 'San Francisco, CA' from multiple job boards. It retrieves the latest 10 results posted within the last 72 hours, specifying 'USA' for Indeed/Glassdoor. The results are returned as a Pandas DataFrame.

import pandas as pd
from jobspy import scrape_jobs

jobs = scrape_jobs(
    site_name=["indeed", "linkedin", "zip_recruiter", "google", "glassdoor"],
    search_term="software engineer",
    location="San Francisco, CA",
    results_wanted=10,
    country_indeed="USA", # Required for Indeed/Glassdoor in many cases
    hours_old=72, # Jobs posted within the last 72 hours
    description_format="markdown",
    verbose=1 # Show warnings and errors
)

if isinstance(jobs, pd.DataFrame):
    print(f"Found {len(jobs)} jobs")
    print(jobs.head())
    # To save to CSV:
    # import csv
    # jobs.to_csv(
    #     "jobs.csv",
    #     quoting=csv.QUOTE_NONNUMERIC,
    #     escapechar="\\",
    #     index=False,
    # )
else:
    print("No jobs found or an error occurred.")

view raw JSON →