Welcome to the Jungle Scraper

Scrape Welcome to the Jungle jobs by role keywords, location, country, contract, remote policy, salary, and other published attributes. Export Welcome to the Jungle job data as JSON, CSV, or spreadsheet files, with titles, pay ranges, offices, descriptions, application links, and employer context. This actor provides a repeatable way to collect public job listings when you need more than a manual search page.

The actor discovers listings from the site's public job sitemaps and reads structured data embedded in each job page. A listing can have several offices, and salary and employer fields depend on what was published. Each result carries a source URL and a stable job ID so later exports can be compared without relying on title alone.

What data can you extract from Welcome to the Jungle?

The default full-detail result has 65 fields. The table describes each output key; examples show the expected shape, while empty values remain null, [], or {} according to the field. Text descriptions retain source HTML so lists and paragraphs are not flattened away.

Field	Description	Example
id	Stable identifier for a listing	fff60753-...
reference	Employer-supplied job reference	ENG-42
name	Published job title	Software Engineer (Senior)
url	Direct source page	https://www.welcometothejungle.com/...
contractType	Contract classification	full_time
remote	Remote-work policy	partial
language	Job language code	en
salaryMin	Published minimum pay	195000
salaryMax	Published maximum pay	250000
salaryCurrency	Currency of pay values	USD
salaryPeriod	Pay period	yearly
experienceLevel	Published experience band	5_TO_7_YEARS
educationLevel	Education level, when given	null
contractDurationMin	Minimum contract duration	null
contractDurationMax	Maximum contract duration	null
publishedAt	Publication timestamp	2026-08-29T00:05:33Z
updatedAt	Last update timestamp	2026-08-29T00:19:58Z
category	Localized occupation category	Tech & Engineering
subcategory	Localized occupation subcategory	Software & Web Development
categoryReference	Stable occupation category reference	tech-engineering-3NjUy
subcategoryReference	Stable occupation subcategory reference	software-web-development-iMzA4
sectors	Employer sector labels	["Software"]
summary	Brief listing summary, if provided	""
offices	All published office objects	[{"city":"New York"}]
city	Primary office city	New York
countryCode	Primary office country	US
organizationName	Employer's published name	[employer]
organizationSlug	Employer's site URL segment	[employer-slug]
organizationLogo	Employer logo URL	https://cdn-images...
organizationEmployees	Published employee count	165
organizationCreationYear	Published founding year	null
organizationLabels	Employer labels	[]
organizationCommitments	Published commitments	[]
organizationIndustry	Published industry label	Software
organizationJobsCount	Employer's published job count	33
description	Full job description in HTML	<ul><li>...</li></ul>
profile	Candidate profile in HTML	<ul><li>...</li></ul>
recruitmentProcess	Hiring steps, if published	null
applyUrl	Direct application destination	https://...
skills	Published skills	["Communication skills"]
tools	Published technical tools	[technology]
benefits	Localized benefit names	[]
benefitDetails	Complete benefit groups	{}
keyMissions	Structured mission list	[]
companySummary	Short employer summary	[summary]
team	Team description	null
startDate	Planned start date	null
atsQuestions	Application questions	[]
applicationFields	Application form requirements	[]
organizationDescription	Employer description in HTML	<p>...</p>
organizationCoverImage	Employer cover image URL	null
organizationRevenue	Published revenue	null
organizationTurnover	Published turnover	null
organizationGdprSetting	Application privacy settings	null
organizationHasExternalAts	External application system flag	true
organizationAutomaticEmail	Automatic email flag	true
organizationWebsite	Employer website URL	https://...
organizationLinkedin	Published social profile	[profile]
organizationTwitter	Published social profile	null
organizationFacebook	Published social profile	null
organizationInstagram	Published social profile	null
organizationAverageAge	Published workforce average age	null
organizationParityMen	Published share of men	null
organizationParityWomen	Published share of women	null
organizationHeadquarter	Headquarters address object	{"city":"San Francisco"}

The office array preserves every location in the source record. Use it when a job is open in several cities: city and countryCode represent only the primary office. Salary values are the numbers published with that job, so compare them only after checking salaryCurrency and salaryPeriod. When the source short summary is empty, the actor derives one from the first 500 characters of the published description.

How to scrape Welcome to the Jungle

1. Enter role words in query, such as software engineer, or leave it blank to discover general listings.
2. Add a city or country and any contract, remote, experience, salary, or occupation filters that matter to the search.
3. Set maxItems to bound the number of saved jobs. Keep includeDetails on if descriptions, application destinations, and employer context are needed.
4. Run the actor and inspect the first few dataset items. Export JSON for nested office and application fields, or CSV for a flat report.

Jobs are saved as they are found. If a run stops before completing its candidate list, items already written to the dataset remain available. For a recurring workflow, save the input alongside each export and compare records by id or url.

Input

Name	Type	Default	Description
maxItems	integer	100	Stop after this many matching jobs; 0 has no result cap.
includeDetails	boolean	true	Add descriptions, application fields, and extended employer data.
websiteCountry	string	en	Select language for localized category and skill labels.
query	string	empty	Match role words in the job URL and page content.
searchOnlyInJobTitle	boolean	false	Restrict the final keyword check to the title.
location	string	empty	Match city, region, or address across published offices.
countryCode	string	empty	Match an office's two-letter country code.
sortBy	string	relevance	Use sitemap order or recently modified candidate pages.
datePosted	string	any	Keep jobs posted in the last 1, 3, 7, 14, or 30 days.
contractType	array	empty	Keep any selected contract type.
remote	array	empty	Keep any selected remote policy.
experienceLevel	string	empty	Minimum experience based on the published band's lower bound.
salaryMin	integer	empty	Keep jobs whose published maximum pay reaches this number.
companySize	string	empty	Filter by published employee-count band.
language	string	empty	Keep jobs written in the supplied language code.
sector	array	empty	Match keywords against published employer sector labels.
occupation	array	empty	Match occupation category references.
proxyConfiguration	object	enabled	Connection settings for public pages.

A realistic input for an initial export:

{
  "query": "software engineer",
  "countryCode": "US",
  "remote": ["partial", "fulltime"],
  "includeDetails": true,
  "maxItems": 50
}

This example is narrower than a title-only search because it checks both office country and remote policy. Start without optional filters if you are exploring which terms appear in the source. Add one filter at a time when a narrow combination returns too few results. For a job title that appears in several languages, a local-language query can expose listings that an English phrase does not.

Output example

The following values come from one job in the successful 30-item validation run. The employer name, employer links, and long HTML text are redacted here; the dataset retains the complete public values.

{
  "id": "fff60753-080e-430d-b9c1-e40217c4fbd6",
  "reference": null,
  "name": "Software Engineer (Senior)",
  "url": "[source job URL]",
  "contractType": "full_time",
  "remote": "unknown",
  "language": "en",
  "salaryMin": 195000,
  "salaryMax": 250000,
  "salaryCurrency": "USD",
  "salaryPeriod": "yearly",
  "experienceLevel": "5_TO_7_YEARS",
  "publishedAt": "2026-08-29T00:05:33Z",
  "category": "Tech & Engineering",
  "subcategory": "Software & Web Development",
  "city": "New York",
  "countryCode": "US",
  "organizationName": "[redacted]",
  "organizationEmployees": 165,
  "organizationJobsCount": 33,
  "description": "[HTML description redacted]",
  "applyUrl": "[application URL redacted]",
  "skills": ["Communication skills", "Collaboration and teamwork"],
  "organizationHasExternalAts": true
}

The validation run returned 30 of 30 requested jobs. Every record had an ID, title, source URL, employer name, publication date, description, and country code. In that sample, published pay bounds appeared in about 73% of records, a primary city in about 77%, and the source short summary was empty throughout, so the actor derived it from the description. Those percentages describe the sample, not a promise for another role or country.

Use cases

• Recruitment analysts can compare published salary ranges and remote policies across roles without opening each listing individually.
• Job seekers can build a focused list of openings and keep the direct application link with the original job page.
• Labor-market researchers can track the mix of contract types, office countries, and occupations in repeated snapshots.
• Employer researchers can connect job demand with published employee counts, sectors, headquarters, and job totals.
• Operations teams can monitor a recurring keyword search and reconcile new or removed IDs between exports.
• Regional researchers can inspect multilingual listings while retaining each job's original language and salary currency.

The data is a snapshot of public pages at run time. A job can be edited, archived, or removed after collection. Keep the publication and update timestamps when comparing snapshots; use the source URL to verify details that affect a decision.

How much does it cost to scrape Welcome to the Jungle?

The current result price is $0.30 per 1,000 saved jobs, with platform usage included in the listed result price. At that rate, 100 jobs cost $0.03, 500 cost $0.15, and 1,000 cost $0.30. These are result-charge examples, not guarantees that the source exposes a requested number of matching jobs. A smaller maxItems makes a first run easier to inspect and limits the number of job pages fetched.

Job pages require residential proxy traffic in the tested environment. The actor obtains one page per candidate to evaluate filters, even when includeDetails is false. Narrow filters can require more page requests than the final result count suggests. If a large export is planned, first run a small sample with the same filters to estimate speed and coverage. The pay-per-result price can change; check the current run price before scheduling a large export.

Tips for best results

Use role words that occur in public job URLs. The actor first checks the URL slug for every word in query, then checks the fetched page content; a role described only inside body text will not be discovered by that search. For more coverage, try a shorter query such as engineer and filter the resulting dataset downstream. A blank query considers all sitemap entries, but broad searches can take much longer because every candidate page must be checked.

Location checks all published office objects, so a multi-office job can match a city other than its primary city value. countryCode offers an exact country test. The location field matches text and does not apply a geographic radius. If a city name is ambiguous, add the country code. websiteCountry changes localized labels where available; it does not restrict jobs to a country. Use language for a job-language filter.

Contract and remote arrays use an any-of match within each field. Across different fields, all supplied conditions must hold. Published remote values include no, partial, punctual, fulltime, and unknown. Salary is checked against the listing's published maximum, with no currency conversion. For pay comparisons across countries, first filter or group by salaryCurrency and salaryPeriod in the exported data.

Use sortBy: "date" when recently modified candidate pages are preferred. That option reads all listed sitemap partitions before fetching jobs and sorts by sitemap modification time. Modification time can differ from publishedAt; datePosted checks the actual publication timestamp after the page is fetched. For routine monitoring, keep the query and filters stable between runs, then compare IDs and update timestamps.

Limits

The public search endpoint returned 403 in probing, including a residential session. This actor uses public sitemap entries and page-embedded job data instead. The site supplied nine job sitemap partitions in the tested index, each with up to 10,000 URLs; that is the discoverable pool at run time, not a complete historical archive. Jobs absent from those sitemaps cannot be found by this actor. Sitemap URLs and page content can change independently, so a listed job may later be unavailable or archived.

The URL prefilter requires every query word to appear in a job slug. This keeps searches practical, but it can miss a relevant listing whose title or description contains a keyword that its URL omits. The relevance option follows sitemap order; it is not ranked search relevance. The date option uses sitemap modification time for candidate order, then reads the publication date for the record. Neither order guarantees all matching records are returned before maxItems is reached.

Detail pages were blocked on datacenter traffic during research and worked on residential traffic. Page availability may vary by region, session, or time. The actor skips a page when its structured job data is unavailable and continues with the next candidate. It does not infer a missing salary, benefit, social profile, or application field. Empty arrays and nulls should be treated as unpublished or unavailable values, not as negative answers.

includeDetails: false removes the extended columns from output but still fetches the job page to apply filters and create the core record. It should be selected for smaller exports, not as a speed or network-cost setting. Large broad searches, narrow filters, and runs with maxItems: 0 can be slow because many candidate pages need to be read.

Use with automated workflows

The actor accepts the same JSON input from a scheduled run or an integration. Give the workflow a bounded result limit and specify the fields needed for the next step. Two example requests are: “Collect up to 50 software engineer jobs in the US and group them by remote policy and salary currency,” and “Find recently posted design jobs in Paris and return source links, publication dates, and direct application destinations.” Review source links before acting on individual jobs.

For recurring monitoring, save the exact input with each dataset and record when it ran. Compare id values across snapshots to find newly observed jobs; compare updatedAt to spot edited records. A missing job in one short, capped run does not establish that it was removed from the site, because candidate order and reachability can vary.

FAQ

Is it legal to collect these listings? The actor reads public job pages. Follow the source site's terms and applicable data-protection rules, and do not misuse personal data. Verify reuse rights before republishing job descriptions or employer media.

Do I need to configure proxies? No extra input is required for the tested setup. The actor uses a datacenter connection for public sitemaps and residential traffic for job pages. The source may change its blocking behavior, so a previously working connection is not a guarantee for future runs.

How fast is a run? Speed depends on the number of candidate pages, filters, and source response times. The 30-item validation run completed successfully, but a country or salary filter may require many more page reads to produce the same number of saved jobs. Test a small, representative input before increasing maxItems.

Can I schedule and monitor collection? Yes. Keep a fixed input and schedule repeated runs. Monitor each run's status and resulting dataset, then compare IDs and timestamps. A failed or partial run should not be used as evidence that a job disappeared.

Can I export to a spreadsheet? Yes. CSV and spreadsheet exports work for the scalar columns. JSON preserves nested offices, headquarters, benefits, and application fields. When importing CSV, keep IDs as text so they are not reformatted.

What does an empty field mean? The public page did not provide that value in the structured record. For example, many employers omit salary or benefits. The actor does not guess, convert currencies, or extract hidden application questions.

Does the location setting support distance search? No. It checks published office text; it does not calculate a radius. Enter a city with a country code to narrow an ambiguous location, or examine the office array after export.

Integrations

Start a run through the platform API, read its default dataset, and send completed results to a database, spreadsheet, or internal report. Webhooks can notify a workflow when a run finishes, and schedules can produce regular snapshots. Store id, url, publishedAt, and updatedAt with downstream records to reconcile later exports. Use JSON when the downstream system needs all office locations or nested application data.

Support

Open an issue on the Issues tab; we reply within 24h and add fields on request.