hh.ru Jobs Scraper – Vacancies, Salaries, Companies

Get hh.ru (headhunter.ru) job search results as clean JSON rows: title, company, city, salary range, experience level, employment type, schedule and requirement/responsibility snippets, for any keyword and region across Russia and the CIS. Built for labour-market analysts, salary-benchmarking teams, recruiters researching companies at scale, and AI agents that need structured job-market data without a login or a browser. Plain HTTPS calls to hh.ru's own public search pages with retries and a silent-failure check, so a scheduled run keeps working when a page hiccups.

What you get

• Every vacancy as one row – id, title, URL, company name/id/URL, city and region id, salary from/to/currency/gross, publication time (UTC), required experience, employment type, work schedule, and requirement/responsibility text snippets.
• Filter by keyword, region and hh.ru's own search facets – multiple keywords in one run, a numeric region id (Russia-wide or any city), experience band, employment type, work schedule and result order, with a per-keyword cap that keeps run cost known up front.
• Company data only, never a contact person – no names, phone numbers or emails are ever extracted, even on vacancies where hh.ru shows a contact card.

Sample row

One row from a real run against hh.ru/search/vacancy?text=python&area=113 (16.09.2026). Raw hh.ru fields not shown here (ad-tracking clickUrl/viewUrl, tags, professionalRoleIds, …) are trimmed for length but present in every row.

{
  "id": 136430777,
  "title": "Преподаватель Python для детей в Новосибирске",
  "url": "https://hh.ru/vacancy/136430777",
  "company_name": "IT-клуб Robi",
  "company_id": 5836965,
  "company_url": "https://hh.ru/employer/5836965",
  "area": "Новосибирск",
  "area_id": 4,
  "salary_from": 20000.0,
  "salary_to": 50000.0,
  "salary_currency": "RUR",
  "salary_gross": false,
  "published_at": "2026-08-20T05:28:57+00:00",
  "experience": "between1And3",
  "employment_type": "PART",
  "schedule": "flexible",
  "requirement_snippet": "Есть опыт преподавания детям / подросткам / студентам. Знания в области программирования, математики и информатики. Хорошие знания языка программирования Python.",
  "responsibility_snippet": "Наличие оборудования для проведения онлайн уроков: компьютер, камера, микрофон, интернет.",
  "is_remote": false,
  "rank_on_page": 0,
  "page": 0,
  "keyword": "python",
  "search_area": "113",
  "scraped_at": "2026-09-16T06:31:08+00:00"
}

Filters

Input	Default	What it does
keywords	["python", "data analyst"]	Search terms; one hh.ru search per keyword.
area	"113" (Russia, all)	hh.ru numeric region id. Common ids: 113 Russia (all), 1 Moscow, 2 Saint Petersburg, 3 Yekaterinburg, 88 Kazan, 76 Rostov-on-Don, 104 Chelyabinsk, 2759 Tashkent, 1002 Minsk, 160 Almaty. Find more via hh.ru's own city picker (the id is in the URL).
maxVacanciesPerKeyword	100	Stop after this many rows per keyword.
searchPeriodDays	empty	Only vacancies published within this many days; empty = hh.ru's own default (no limit).
experience	empty (any)	noExperience, between1And3, between3And6, moreThan6.
employment	empty (any)	full, part, project, volunteer, probation.
schedule	empty (any)	fullDay, shift, flexible, remote, flyInFlyOut.
orderBy	relevance	relevance, publication_time, salary_desc, salary_asc.

An unrecognised area id never silently scrapes the wrong scope: hh.ru is asked, and if it doesn't echo the id back as accepted, the keyword becomes one uncharged error row (keyword, area, error, errorMessage) instead of a wrong or empty result set, and the run still finishes.

Pricing

Pay per vacancy, no subscription: $1 per 1,000 vacancies. A default run (up to 200 rows across the two default keywords) costs at most $0.20. Set a maximum charge on the run and the Actor stops cleanly when it is reached, charging only for rows that were actually saved.

Details

• Source: hh.ru's public search pages (hh.ru/search/vacancy), reading the HH-Lux-InitialState JSON embedded in the page HTML (enable_snippets=true turns on the requirement/responsibility text). No authentication, no proxies, no browser.
• Why HTML and not the official API: api.hh.ru/vacancies returned HTTP 403 forbidden in testing, both with a browser user agent and with a descriptive API-style one; the public search page returns the same underlying data at HTTP 200.
• No contact data: only public listing and company information is extracted. Contact-person names, phone numbers and emails (hh.ru shows these on some vacancies) are dropped before the row is built, never included even in the raw-field dump.
• is_remote: a best-effort flag, true only when hh.ru's own schedule is exactly "remote". Hybrid or partial-remote arrangements are not reliably distinguishable from the search page and are left null/false rather than guessed.
• Reliability: 429 and 5xx responses are retried with exponential backoff (5 tries), a 200 without a parseable HH-Lux-InitialState block counts as a failure, and one bad keyword never stops the run: it becomes an error row and the rest continues. A run fails only when it produced no rows and hit a genuine fetch failure (network, rate limit, blocked, not found, or an unexpected page layout); an invalid area or a keyword with zero matches are both handled outcomes, not failures.
• Run stats: the STATS record in the run's key-value store holds request and error counts per category (network, rate_limit, blocked, not_found, other).
• Pagination: up to 100 vacancies per request page, following hh.ru's own last-page count until maxVacanciesPerKeyword is reached or pages run out.
• Output: one dataset row per vacancy with the columns above first and every other raw hh.ru search field after them (minus contact/address/internal-manager blocks, which are never included). Export as JSON, CSV or Excel, fetch through the Apify API, or schedule runs and pipe them into Google Sheets, Make, Zapier or your own code.

Not affiliated with HeadHunter (hh.ru).