HeadHunter (hh.ru) Jobs Scraper

Extract job vacancies — with full descriptions, employers, salaries and regions — from HeadHunter, the largest job site in Russia, Kazakhstan, Uzbekistan and Belarus (hh.ru, hh.kz, hh.uz, rabota.by).

Why use this actor

• No account, no login, no API key. HeadHunter's own developer API refuses anonymous traffic; this actor reads the public site directly, so there is nothing to register or renew.
• Four countries from one actor. hh.ru, hh.kz, hh.uz and rabota.by all run on one search index — pick the site, then filter to any country, region or city.
• The whole ad, not just the card. Every result can carry the complete vacancy text, key skills, employer profile, exact publication date and salary breakdown (gross/net, per month or per shift).
• Filters that actually apply. Region, experience, employment type, schedule (including remote-only), minimum salary, "posted in the last 24 hours", and sort-by-date for monitoring runs. Region values are verified against HeadHunter's own region list before the run starts — an unrecognised region is reported as an error instead of quietly returning nationwide results.
• Stable JSON output suitable for pipelines, databases and spreadsheets, with automatic retries built in.
• Runs on Apify's schedule; export to JSON, CSV or Excel.

How it works

1. You give it one or more search keywords (Russian or English), plus any filters you want — region, salary, schedule, how recently the ad was posted.
2. It runs each keyword as its own search and walks through the result pages, collecting every distinct vacancy until it reaches your limit or the site runs out of results.
3. For each vacancy it optionally opens the full ad to add the complete description, key skills and employer details.
4. You can also hand it specific vacancy URLs or ids to scrape directly, skipping search entirely — the two modes can be combined in one run.
5. Everything lands in one dataset: a summary row per search, then one row per vacancy.

No scrapers, browsers, or blocks to babysit.

Input

{
  "site": "hh.ru",
  "keywords": ["python"],
  "areas": ["1"],
  "maxItemsPerQuery": 100,
  "includeVacancyDetails": true,
  "searchFields": [],
  "experience": "",
  "employment": [],
  "schedule": ["remote"],
  "searchPeriod": "",
  "onlyWithSalary": false,
  "orderBy": "publication_time",
  "vacancyUrls": ["https://hh.ru/vacancy/136131345"],
  "maxConcurrency": 4,
  "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Field	Type	Description
site	string	Which portal to read: hh.ru (Russia), hh.kz (Kazakhstan), hh.uz (Uzbekistan), rabota.by (Belarus). All four share one index, so this sets the site context (locale, currency, vacancy URLs) — use areas to actually filter by country.
keywords	array	One search per entry; each gets its own summary row. Russian or English. Leave empty to browse everything matching your filters.
areas	array	Region ids or names: 113 Russia, 1 Moscow, 2 Saint Petersburg, 40 Kazakhstan, 160 Almaty, 159 Astana, 97 Uzbekistan, 2759 Tashkent, 16 Belarus, 1002 Minsk. Names as HeadHunter spells them (Алматы) work too.
vacancyUrls	array	Scrape specific ads directly. Full URL (https://hh.ru/vacancy/136131345) or bare id. Can be combined with a keyword search.
maxItemsPerQuery	integer	Stop after this many vacancies per search. 0 = unlimited (up to the site's own 2000-per-search ceiling).
includeVacancyDetails	boolean	Fetch each ad's full text and employer details (one extra request per vacancy). Turn off for a fast listing-only crawl.
searchFields	array	Match the keyword only in name (title), company_name, or description. Empty = search everywhere.
experience	string	noExperience, between1And3, between3And6, moreThan6.
employment	array	full, part, project, volunteer, probation.
schedule	array	fullDay, shift, flexible, remote, flyInFlyOut.
searchPeriod	string	Only ads posted in the last 1, 3, 7 or 30 days.
onlyWithSalary	boolean	Skip ads that don't state pay.
salary	integer	Minimum salary, in the site's own currency.
orderBy	string	relevance, publication_time, salary_desc, salary_asc.
maxConcurrency	integer	Requests in flight at once (1–16).
proxyConfiguration	object	Residential recommended for scheduled runs.

Output

Summary row — one per search

{
  "_input": "hh.ru | text='python' | area=Москва",
  "_source": "S1-search-state",
  "_scrapedAt": "2026-08-12T16:44:20Z",
  "recordType": "SEARCH_SUMMARY",
  "site": "hh.ru",
  "siteName": "Russia (hh.ru)",
  "totalResults": 1312,
  "returnedCount": 2,
  "pagesFetched": 1,
  "areasApplied": [
    { "input": "1", "id": "1", "name": "Москва", "status": "validated" }
  ],
  "resultCap": 2000,
  "resultCapReached": false,
  "requestParams": [
    { "key": "page", "value": "0" },
    { "key": "items_on_page", "value": "100" },
    "... 3 more"
  ]
}

Field	Type	Description
totalResults	integer	How many vacancies HeadHunter says match the search.
returnedCount	integer	How many this run actually collected.
pagesFetched	integer	Result pages read.
areasApplied	array	Each requested region, its resolved id and name, and whether it was validated (checked against HeadHunter's region list) or unverified.
resultCap / resultCapReached	integer / boolean	The site serves at most 2000 results per search; resultCapReached is true when more matches exist than can be paged through. Narrow the search to reach the rest.
_warning	string	Present when something needs your attention, e.g. the 2000 ceiling was hit.

Vacancy row — one per job

{
  "_input": "hh.ru | text='python' | area=Москва",
  "_source": "S1-search-state+S2-vacancy-detail",
  "_scrapedAt": "2026-08-12T16:44:20Z",
  "recordType": "VACANCY",
  "site": "hh.ru",
  "vacancyId": "131031614",
  "vacancyUrl": "https://hh.ru/vacancy/131031614",
  "searchContext": { "keyword": "python", "page": 0, "rank": 1 },
  "name": "Cпециалист по автоматизации технических процессов",
  "company": {
    "id": 9498112,
    "name": "Яндекс Крауд",
    "visibleName": "Яндекс Крауд: Бэк-офис",
    "accreditedITEmployer": false,
    "@trusted": true
  },
  "compensation": {
    "from": 78000, "to": 78000, "currencyCode": "RUR", "gross": false, "mode": "MONTH"
  },
  "area": { "@id": 1, "name": "Москва", "path": ".113.232.1." },
  "publicationTime": { "@timestamp": 1786440622, "$": "2026-08-11T12:30:22.769+03:00" },
  "@workSchedule": "remote",
  "workFormats": [{ "workFormatsElement": ["REMOTE"] }],
  "workScheduleByDays": [{ "workScheduleByDaysElement": ["FIVE_ON_TWO_OFF"] }],
  "workingHours": [{ "workingHoursElement": ["HOURS_8"] }],
  "workExperience": "between1And3",
  "employmentForm": "FULL",
  "responsesCount": 0,
  "totalResponsesCount": 2616,
  "links": {
    "desktop": "https://hh.ru/vacancy/131031614",
    "mobile": "https://m.hh.ru/vacancy/131031614"
  },
  "vacancyDetails": {
    "vacancyId": 131031614,
    "name": "Cпециалист по автоматизации технических процессов",
    "description": "<p>Откликайтесь, если самостоятельно писали код и составляли SQL-запросы, увлечены технологиями и хотите строить карьеру в динамичной IT-отрасли</p> <p><strong>Какие задачи вас ждут</strong></p> <ul> <li> <p>Создавать и дорабатывать технические процессы, используя Groovy, Python и внутренние инструменты</p> </li> ...",
    "publicationDate": "2026-08-11T12:30:22.769+03:00",
    "workFormats": ["REMOTE"],
    "keySkills": null,
    "address": { "mapData": null, "metroStations": null }
  }
}

Field	Type	Description
vacancyId / vacancyUrl	string	Id and canonical link for the ad.
name	string	Job title.
company	object	Employer: id, name, display name, logos, IT-accreditation flag, trusted flag.
compensation	object	Salary range, currency, whether it's gross or net, and the period (MONTH, shift, etc.). Absent or noCompensation when pay isn't stated.
area	object	Region id, name, and its position in HeadHunter's region tree.
publicationTime	object	When the ad was published, as a timestamp and an ISO date.
workFormats, workScheduleByDays, workingHours, workExperience, employmentForm	mixed	Remote/on-site, shift pattern, hours per day, required experience, employment type.
responsesCount / totalResponsesCount	integer	How many people have applied — useful as a competition signal.
searchContext	object	Which keyword, page and rank this row came from. Absent in direct-URL mode.
vacancyDetails	object	The full ad: complete HTML description, keySkills, publicationDate, address with metro stations, employer branding, and ~60 more fields. null when includeVacancyDetails is off.

Rows are passed through with HeadHunter's own field names, so new fields upstream flow straight into your dataset instead of being dropped.

Error row

Every input produces at least one row. Failures are reported, never silently skipped:

{
  "_input": "https://hh.ru/vacancy/1",
  "_source": "none",
  "_scrapedAt": "2026-08-12T16:41:52Z",
  "recordType": "ERROR",
  "site": "hh.ru",
  "_error": "not_found",
  "_errorDetail": "vacancy archived or removed"
}

Notes and limits

• 2000 results per search, set by the site itself. Any search reporting more matches than that hits the ceiling — the summary row flags it in resultCapReached and _warning. Split by region, by searchPeriod, or by narrower keywords to get the rest.
• Choosing a site does not filter by country. All four portals share one index, so hh.kz alone still returns Russian vacancies. Use areas for real country/city filtering.
• Unknown regions are refused, not guessed. HeadHunter answers an unrecognised region id with a normal-looking page of unfiltered results, so this actor checks every region up front and returns an invalid_area error instead of shipping nationwide data labelled as a city search.
• Vacancies expire. Ads archived between the search and the detail fetch are marked with a _detailError on that row; the rest of the run is unaffected.
• HeadHunter mixes a few sponsored placements into results. Most are ordinary job ads and are returned normally. The occasional promotional slot that isn't a job at all is marked not_a_vacancy on its row instead of being passed off as a vacancy.
• Residential proxy is recommended for scheduled runs.