Welcome to the Jungle Jobs Scraper & New Job Monitor

Get Welcome to the Jungle jobs as clean JSON: job title, company with size, founding year and sector, contract type, remote policy, salary as numbers (min, max, currency, period, and the yearly minimum and maximum), experience, office city and country with coordinates, date published and the link. Search by keyword, company, country, city and radius, contract, remote, experience, company size, sector, job category, education, minimum yearly salary and date posted. Turn on Monitor mode and scheduled runs return only jobs published since the last check, for a job alert, a Slack bot or a hiring-signal lead list. Turn on job page details to add the full description, candidate profile, skills, tools, start date, contract length, number of hiring steps, and the company's website, LinkedIn, revenue and headquarters.

On 2026-10-21 Welcome to the Jungle listed 89,976 jobs: 59,190 with an office in France, 18,264 in the US, 5,471 in the UK, 1,858 in Canada, 1,149 in Germany, 895 in Spain and more. 28,426 of them state a salary amount.

• All the jobs, not just the first 1,000. In newest-first order (the default), remote data or all of Paris comes back in full, up to maxResultsPerSearch (50,000).
• Salary you can sort and filter. salaryMin, salaryMax, salaryCurrency, salaryPeriod as stated in the ad, plus salaryYearlyMin and salaryYearlyMax so monthly, daily and yearly pay can be compared. Filter with minYearlySalary, salaryCurrency and onlyWithSalary.
• Company data in every row, without extra cost. Employees, founding year, sector, industry, logo and company page come with the basic row. With job page details you also get the company website, LinkedIn, revenue, average age, number of open jobs and headquarters.
• Clean text. Description and candidate profile keep their line breaks and lists; HTML is removed. Skills, tools, missions and benefits are lists.
• New job monitor. The first run only remembers what matches now and is free. Later runs return only jobs published since, including jobs the site indexes a few hours late. When nothing is new, the run costs one small check fee instead of a full search.
• Each job is charged once. Overlapping keywords and companies return each job once in a run, and a monitor does not return a job again in a later run, even under another search of the same watch. A job that closed between the search and its job page is skipped, not sold, and the next job takes its place.
• Free rows say why. No match, an unknown company name, nothing new, a refused or unreadable request, a search that has more jobs than you asked for, and a run that reached its maximum charge each come back as a free row with a plain-English note.

Unofficial. Reads the public job search and the public job pages of welcometothejungle.com, the same data a visitor sees without logging in. It does not log in, does not apply to jobs and does not solve or bypass check pages. It returns the company, not the people: there are no recruiter or contact person columns, the hiring process is returned as a number of steps only, and phone numbers and e-mail addresses in text are replaced with [phone removed] and [email removed].

What you get

One row per job. Example (job page details on, description, candidateProfile, companyDescription and the logo link shortened here):

{
  "status": "ok",
  "changeType": null,
  "jobId": "3fd70968-dccf-47ad-852f-1d6d1966eb68",
  "jobSlug": "migration-factory-orchestrator_paris",
  "title": "Migration Factory Orchestrator",
  "url": "https://www.welcometothejungle.com/en/companies/inqom/jobs/migration-factory-orchestrator_paris",
  "publishedAt": "2026-10-20T15:57:33.000Z",
  "company": "Inqom",
  "companySlug": "inqom",
  "companyUrl": "https://www.welcometothejungle.com/en/companies/inqom",
  "companyLogoUrl": "https://cdn-images.welcometothejungle.com/...",
  "companyEmployees": 112,
  "companyFoundedYear": 2016,
  "companySummary": "Logiciel comptable et fiscal SaaS pour cabinets et directions financières.",
  "companySectors": ["Artificial Intelligence / Machine Learning", "FinTech / InsurTech", "SaaS / Cloud Services"],
  "companyIndustry": "Tech",
  "category": "Tech & Engineering",
  "subCategory": "Data / Business Intelligence",
  "profession": "Data Engineer",
  "contractType": "full_time",
  "contractDurationMinMonths": null,
  "contractDurationMaxMonths": null,
  "remote": "fulltime",
  "hasRemote": true,
  "salaryMin": 51000,
  "salaryMax": 53000,
  "salaryCurrency": "EUR",
  "salaryPeriod": "yearly",
  "salaryYearlyMin": 51000,
  "salaryYearlyMax": 53000,
  "experienceYearsMin": 4,
  "educationLevel": null,
  "language": "fr",
  "city": "Paris",
  "region": "Ile-de-France",
  "country": "France",
  "countryCode": "FR",
  "offices": [{ "city": "Paris", "region": "Ile-de-France", "country": "France", "countryCode": "FR" }],
  "latitude": 48.86764,
  "longitude": 2.32294,
  "summary": "Rejoins Inqom, une entreprise en pleine croissance dans le secteur de la technologie...",
  "keyMissions": ["Cartographier le parcours de bout en bout, formaliser les étapes, les prérequis, les contrôles et les responsabilités.", "..."],
  "benefits": ["Flexible working hours", "Cooptation bonus", "Team building", "Professional development plan", "..."],
  "sponsored": false,
  "description": "...Le poste couvre les parcours et lots standardisables...",
  "candidateProfile": "- Expérience confirmée en opérations, industrialisation de processus...",
  "applyUrl": null,
  "applyOnSite": true,
  "skills": ["Project management", "Communication skills", "Automation tools", "Mechanical skills", "Team management"],
  "tools": ["N8n"],
  "recruitmentSteps": 5,
  "startDate": null,
  "updatedAt": "2026-10-20T16:13:17.480001Z",
  "officeAddress": "8 Rue Royale, 75008 Paris, France",
  "officePostalCode": "75008",
  "canonicalUrl": "https://www.welcometothejungle.com/fr/companies/inqom/jobs/migration-factory-orchestrator_paris",
  "companyWebsite": "http://www.inqom.com",
  "companyLinkedin": "inqomapp",
  "companyIndustryTags": ["Artificial Intelligence / Machine Learning", "FinTech / InsurTech", "SaaS / Cloud Services"],
  "companyRevenue": null,
  "companyAverageAge": 37,
  "companyOpenJobs": 19,
  "companyHeadquarters": "8, Rue Royale, Paris",
  "companyHeadquartersCity": "Paris",
  "companyHeadquartersCountryCode": "FR",
  "companyDescription": "Fondé en 2016, Inqom est un logiciel moderne de production comptable et fiscale en SaaS...",
  "detailsStatus": "ok",
  "searchQuery": "data",
  "searchCompany": null,
  "watchName": null,
  "checkedAt": "2026-10-21T06:00:00.000Z"
}

Without job page details, a row has every column up to sponsored plus salaryYearlyMax, detailsStatus (null), dataNotes, searchQuery, searchCompany, watchName and checkedAt.

• contractType: full_time (permanent), part_time, temporary, internship, apprenticeship, freelance, vie, idv, graduate_program, volunteer, other.
• remote: fulltime (fully remote), partial (hybrid), punctual (occasional), no (on site), unknown.
• salaryYearlyMin is the site's own yearly figure; salaryYearlyMax applies the same factor to salaryMax. When the ad gives only a maximum, salaryYearlyMax uses the usual factor of its period (×12 monthly, ×218 daily) and salaryYearlyMin stays null. Both are null when the ad states no salary.
• dataNotes lists short notes when the site's own data contradicts itself (for example a minimum salary above the maximum, a minimum contract length above the maximum, or a publication time in the future). The values are shown exactly as listed, not swapped. It is null otherwise.
• changeType is new in monitor mode and null otherwise.
• detailsStatus is ok when the job page was read, unreadable when it could not be (the row is then returned without details and charged as a job without details), and null when details were not asked for.

Input

• searchQueries: what you would type into the site's search box (data engineer, product manager, comptable). Each keyword is one search. Empty = every job that matches the filters.
• companies: company names from the company page address (doctolib, back-market) or the company page URL. Each company is searched with each keyword.
• searchInTitleOnly: a keyword must be in the job title.
• siteLanguage: language of links and of category, sector and country names (en, fr, es, cs, sk).
• Filters: countries (FR, US, GB ...), cities and radiusKm, contractTypes, remote, experienceLevels, companySizes, educationLevels, jobLanguages, sectors, categories, minYearlySalary, salaryCurrency, onlyWithSalary, postedWithinDays.
• sortBy: date (newest first, reads past 1,000) or relevance (the site's ranking, up to 1,000 per search).
• maxResultsPerSearch: default 100.
• includeJobDetails: read each job page for the extra columns.
• onlyNew, watchName, resetMonitoringState: monitor mode.

Up to 20 searches (keywords times companies) per run.

Examples

Remote data jobs in France, newest first:

{ "searchQueries": ["data engineer", "data analyst"], "countries": ["FR"], "remote": ["fulltime", "partial"], "maxResultsPerSearch": 200 }

Every internship in Paris and 30 km around it, posted in the last 7 days:

{ "cities": ["Paris"], "radiusKm": 30, "contractTypes": ["internship"], "postedWithinDays": 7, "maxResultsPerSearch": 2000 }

All open jobs at two companies, with job page details:

{ "companies": ["doctolib", "qonto"], "includeJobDetails": true, "maxResultsPerSearch": 500 }

Daily alert for well-paid product jobs in the US (schedule it once a day):

{ "searchQueries": ["product manager"], "countries": ["US"], "minYearlySalary": 150000, "salaryCurrency": "USD", "onlyNew": true, "watchName": "us-pm-150k" }

Pricing

Pay per event:

• Job returned: per job row without job page details.
• Job with details: per job row with job page details (includeJobDetails), instead of Job returned. A row whose job page could not be read is charged as Job returned.
• Monitoring check: in monitor mode, once per search when the check found no new job. Not charged on the first (baseline) run, when new jobs are returned, or when nothing could be read.

Free rows are never charged. If you set a maximum total charge for a run, the run stops before it would go over it and says so in a free row; job pages are only read for rows that fit.

Monitor mode, step by step

1. First run: remembers the newest jobs that match each search (up to 1,000 per search) and returns a free baseline-saved row. Nothing is charged.
2. Later runs: read the jobs published since shortly before the last check (2 days of slack, so jobs the site indexes late are still caught) and return the ones this watch has not seen as new, up to maxResultsPerSearch. When there is none, the run returns a free no-new-jobs row and charges one monitoring check. A job returned under one search of a watch is remembered for the whole watch, so it is not returned or charged again under its other searches.
3. Jobs not returned because of maxResultsPerSearch or the charge limit are not remembered, so the next run returns them.
4. Nothing is judged from a search that could not be read (nothing returned, remembered or charged); the next run checks it again.

Use watchName to keep separate monitors apart, and resetMonitoringState to take a new baseline.

Good to know

• relevance order follows the site's own ranking, which stops at 1,000 jobs per search; when more match, a free row says so. date order has no such stop.
• A monitor run reads up to 5,000 jobs per search. If more new jobs than that arrive between two checks, a free row says so and the next run continues.
• About 1 in 3 jobs states a salary amount. Without one, all salary columns are null; nothing is estimated.
• radiusKm needs exactly one city, and the city must be written as the site writes it (Paris, Lyon, London, New York).
• Most jobs are in French or English. siteLanguage changes the language of links and labels, not of the ad itself; use jobLanguages to pick the ad language.
• Names in free text (for example a signature at the end of a description) are not recognised; phone numbers and e-mail addresses are removed.

Support

Questions, a field you need, or a search that does not work: open an issue on the Issues tab of this Actor.

Thanks for using this Actor

We build these tools for people who run them every day, and we improve them from what users tell us.

• Missing a field, or need another filter? Tell us in the Issues tab. If the data is there, we add it.
• Found a bug or a wrong value? Post the run ID in the Issues tab. Wrong data is the thing we fix first.

If this Actor saved you time, a short review helps other people find it.