CareerCross Scraper — Bilingual Japan Job Listings

Scrape job listings from CareerCross, Japan's bilingual (English/Japanese) job board for English-speaking talent in Japan, global companies hiring locally, and inbound talent. Pull title, company, salary range, location, employment type, English/Japanese level, visa status, industry, and job categories — flat record per posting.

What it returns

Field	Description
job_id, url	CareerCross numeric job ID and canonical detail URL
title	Listing title (Japanese, English, or bilingual)
company_name, company_legal_name, company_description	Hiring company display name, legal name, and description block
location, nearest_station, prefecture	Address line, advertised station, and parsed prefecture name
employment_type	Mapped label: Permanent Full-time, Part-time, Contract, Temp to Perm, Freelance, etc.
experience_required_label, experience_required_months	Original label plus a numeric months floor (0 / 12 / 36 / 72 / 120)
education_requirement	e.g. "Bachelor's Degree", "High School"
salary_min, salary_max, salary_currency, salary_period, salary_label	Numeric range from data-salary-* attrs (with label-fallback parsing), currency JPY, period inferred from the label
english_level, japanese_level	Required language proficiency strings
visa_sponsorship, visa_status	Boolean (when detectable from the visa-status text) plus the raw status string
industry, job_categories	Industry label + breadcrumb categories ("Top > Sub" strings)
career_level, work_hours	Career stage label and work-hours string
description_html, description_text, required_skills	Description in HTML and stripped text + bullet-array of required skills
date_posted	Last-updated timestamp (ISO-8601, JST +09:00)
language, source_search_url, scraped_at	Page locale (en / ja), originating search-result page (when applicable), and emit timestamp

Three input modes

The actor accepts one of three input patterns. Pick the mode and supply the matching field.

1. mode: "search" — keyword search (default)

Best for browsing or topic discovery. The actor submits a keyword search, follows the redirect to a fresh search-result page, and walks pagination collecting detail records up to maxItems.

{
  "mode": "search",
  "keyword": "engineer english",
  "language": "en",
  "maxItems": 50
}

language toggles between /en/ (English UI) and /jp/ (Japanese UI). Both locales use the same selectors — pick whichever matches your downstream processing.

2. mode: "urls" — direct URLs

Use when you already have CareerCross URLs to scrape. Accepts both detail pages and search-result pages.

{
  "mode": "urls",
  "startUrls": [
    { "url": "https://www.careercross.com/en/job/detail-1589338" },
    "https://www.careercross.com/en/job-search/result/64776616?page=1"
  ],
  "maxItems": 100
}

Detail URLs produce one record each. Search-result URLs walk pagination from the supplied page forward, same as search mode.

3. mode: "ids" — numeric job IDs

Use when you have a list of CareerCross job IDs from another system.

{
  "mode": "ids",
  "jobIds": ["1589338", "1577447", "1526389"],
  "language": "en",
  "maxItems": 100
}

Each ID becomes https://www.careercross.com/<lang>/job/detail-<id>.

Facet filters (optional)

Available in every mode. Each filter is a multi-select array — narrows results to rows that verifiably match at least one selected value; leave empty for the previous unfiltered behaviour.

• prefectures — keep only rows whose parsed prefecture is one of the selected Japan prefectures (e.g. ["Tokyo", "Osaka"]).
• experienceLevels — keep only rows whose experience-requirement bucket (the site's own code) matches one of the selected levels ("0"=no experience required … "4"=10+ years).
• salaryBands — keep only rows whose annual-JPY salary range overlaps one of the selected bands (under_4m, 4m_6m, 6m_8m, 8m_10m, 10m_plus). Rows quoted per-month/per-hour are excluded when this filter is set.

A row with unrecognised/unverifiable data for an active filter is dropped, never passed through as if unfiltered.

Output sample

{
  "job_id": "1589338",
  "url": "https://www.careercross.com/en/job/detail-1589338",
  "title": "Resort Operations Staff (Fukushima/Inawashiro) — English required",
  "company_name": "DMCaizu Co., Ltd.",
  "company_legal_name": "DMC aizu Co., Ltd.,",
  "location": "Fukushima Prefecture, Yama-gun Inawashiro-machi",
  "nearest_station": "Banetusai Line Station",
  "prefecture": "Fukushima",
  "employment_type": "Permanent Full-time",
  "experience_required_label": "Over 1 year",
  "experience_required_months": 12,
  "education_requirement": "Bachelor's Degree",
  "salary_min": 4000000,
  "salary_max": 6000000,
  "salary_currency": "JPY",
  "salary_period": "year",
  "salary_label": "4 million yen ~ 6 million yen",
  "english_level": "Business Level (Amount Used: English usage about 25%)",
  "japanese_level": "Fluent",
  "visa_sponsorship": false,
  "visa_status": "Permission to work in Japan required",
  "industry": "Tourism",
  "job_categories": [
    "Education Jobs > Instructor, Teacher, Coach Jobs",
    "Hotel, Travel and Hospitality Jobs > Reservation Agent, Guest Service Agent Jobs"
  ],
  "career_level": "Mid Career",
  "work_hours": "8:00-17:00 (1hr break)",
  "date_posted": "2026-04-25T12:02:17+09:00",
  "language": "en",
  "source_search_url": "",
  "scraped_at": "2026-04-25T14:51:00.000Z"
}

Pricing

Pay-per-event:

• $0.10 per actor start
• $0.003 per data record

Standard pay-per-event pricing on the data event, reflecting the bilingual Japan-niche specialty data. Most search runs cost the start fee plus a few cents in records — a maxItems: 100 run that returns 100 records is $0.40 total.

Notes

• Pagination on search-result pages is ?page=N; the actor stops at the page boundary or at maxItems, whichever comes first.
• Job postings update frequently — re-run with the same keyword/sid daily for fresh listings, or pin to a specific result/<sid> URL via mode='urls' to re-poll the same canonical search.

— BowTiedRaccoon

---

Further reading: Job Postings Data: How to Get Listings From 12 Boards Across 10 Markets