How to Scrape Upwork: A Comprehensive Technical Guide

from playwright.sync_api import sync_playwright

def scrape_upwork():
    with sync_playwright() as p:
        # Launching with stealth-like settings or residential proxies is advised
        browser = p.chromium.launch(headless=True)
        context = browser.new_context(user_agent='Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36')
        page = context.new_page()
        
        # Navigate to a public search page
        page.goto('https://www.upwork.com/freelance-jobs/web-development/')
        
        # Wait for the dynamic content to load
        page.wait_for_selector('[data-test="job-tile-list"]')
        
        jobs = page.query_selector_all('.job-tile')
        for job in jobs:
            title_el = job.query_selector('.job-title')
            if title_el:
                print({'title': title_el.inner_text()})
        
        browser.close()

scrape_upwork()

import scrapy

class UpworkSpider(scrapy.Spider):
    name = 'upwork'
    start_urls = ['https://www.upwork.com/freelance-jobs/data-science/']

    def parse(self, response):
        # Upwork uses dynamic class names frequently
        for job in response.css('.job-tile'):
            yield {
                'title': job.css('.job-title::text').get(),
                'budget': job.css('.job-type::text').get(),
                'link': response.urljoin(job.css('a::attr(href)').get())
            }

        # Pagination handling
        next_page = response.css('button.next-page-btn::attr(href)').get()
        if next_page:
            yield response.follow(next_page, self.parse)

const puppeteer = require('puppeteer');

(async () => {
  const browser = await puppeteer.launch({ headless: true });
  const page = await browser.newPage();
  
  // Emulating a real user
  await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36');
  
  await page.goto('https://www.upwork.com/freelance-jobs/');
  
  // Wait for the job feed to render
  await page.waitForSelector('.job-tile');
  
  const jobs = await page.evaluate(() => {
    return Array.from(document.querySelectorAll('.job-tile')).map(el => ({
      title: el.querySelector('.job-title')?.innerText.trim(),
      posted: el.querySelector('.posted-date')?.innerText.trim()
    }));
  });
  
  console.log(jobs);
  await browser.close();
})();