JobsDB Scraper — Hong Kong & Thailand | Salary, Descriptions & Contact Data

Extract structured job listings from JobsDB Hong Kong (hk.jobsdb.com) and JobsDB Thailand (th.jobsdb.com) — the SEEK-platform job boards covering two of Asia's largest hiring markets. No proxy required. No API key needed. Runs entirely in the cloud.

---

What data can you get?

Each job record includes:

• Title, company, location, work type, work arrangement (on-site / hybrid / remote)
• Salary — raw label + parsed salaryMin / salaryMax / salaryCurrency / salaryPeriod
• Full job description — HTML, plain text, and Markdown (when fetchDetails is enabled)
• Classification & sub-classification — industry category IDs and labels
• Company intelligence — company logo, banner image, company profile ID
• Job score (0–5) — transparent signal based on salary completeness, description quality, posting recency, and contact presence
• Tags — EARLY_APPLICANT, EXPIRING_SOON, and other listing signals
• Bullet points — key highlights shown on listing cards
• Posted date, expiry date — ISO timestamps

---

Use cases

Recruitment intelligence

Track job demand by role, location, and industry. Compare hiring trends across HK and TH, monitor which classifications are growing, and benchmark posting volume over time.

Salary benchmarking

Collect salary labels across roles and locations for compensation research. The actor parses salary ranges into structured salaryMin / salaryMax fields for direct analysis — no manual parsing needed.

Market monitoring

Run daily or hourly with skipReposts: true and maxAgeMinutes to receive only fresh listings. Combine with keyword filters to track specific roles or companies.

AI agent and LLM pipelines

Use compactOutput: true to produce lean payloads. Sort by jobScore to surface high-quality, information-rich listings for downstream processing.

---

Input parameters

Parameter	Type	Default	Description
country	select	HK	Market to scrape: HK (Hong Kong) or TH (Thailand).
searchQuery	string	—	Keyword, role title, or skill. Leave empty to browse all jobs. Pass a JSON array for multi-query mode: ["python developer", "data analyst"] — each query runs separately and results are merged and deduplicated.
location	string	—	City, district, or region. E.g. Kowloon, Bangkok. Leave empty for all locations.
workType	select	All	Filter by: Full time, Part time, Contract/Temp, Casual.
classificationId	integer	—	Industry category ID. E.g. 6281 for IT, 6251 for Accounting.
subClassificationId	integer	—	Sub-category ID. Requires classificationId.
sortMode	select	ListedDate	Sort by date posted (newest first) or relevance.
daysOld	integer	—	Restrict to jobs posted within N days (max 30).
includeKeywords	string	—	Comma-separated keywords that must appear anywhere in a listing (title, company, description, etc.). All terms must match. Example: python, remote
excludeKeywords	string	—	Comma-separated keywords that must not appear. Any match discards the listing. Example: intern, unpaid
fromDate	string	—	Keep only jobs posted on or after this date. Format: YYYY-MM-DD.
toDate	string	—	Keep only jobs posted on or before this date. Format: YYYY-MM-DD.
maxAgeMinutes	integer	—	Keep only jobs posted within the last N minutes. Useful for near-real-time monitoring.
fetchDetails	boolean	false	Fetch full description, salary, and company data per job. Slower but richer output.
compactOutput	boolean	false	Omit description fields. Job score still runs. Ideal for LLM pipelines.
skipReposts	boolean	false	Skip listings already seen in a previous run (90-day memory).
startUrls	array	—	Specific JobsDB job URLs to scrape directly, bypassing search.
maxResults	integer	5	Maximum number of listings to return. Post-filters apply after fetch — you may receive fewer results than this limit.

---

Output schema

Each result item contains:

Field	Description
id	JobsDB listing ID
url	Direct link to the job posting
title	Job title
companyName	Employer name
advertiserId	SEEK advertiser ID
companyId	SEEK company profile ID (when available)
companyLogo	Company logo URL (when available)
companyBannerUrl	Company banner image URL (when available)
companyIsVerified	Whether the listing is branded / verified
location	Location label
countryCode	HK or TH
workType	Employment type (Full time, Part time, etc.)
workArrangement	On-site, Hybrid, or Remote
displayType	Listing tier (standard, branded, etc.)
jobAdType	ORGANIC or SPONSORED
isFeatured	Featured listing flag
listingDate	ISO 8601 posting date
listingDateDisplay	Human-readable age ("3d ago")
expiryDate	ISO 8601 expiry date
salaryLabel	Raw salary text from listing
salaryMin	Parsed minimum salary amount
salaryMax	Parsed maximum salary amount
salaryCurrency	HKD or THB
salaryPeriod	month, year, hour, etc.
classification	Industry category label
classificationId	Industry category ID
subClassification	Sub-category label
subClassificationId	Sub-category ID
roleId	Role slug (e.g. software-engineer)
tags	Listing tags (e.g. Be an early applicant)
bulletPoints	Key highlight bullet points from card
teaser	Short listing summary
descriptionHtml	Full description HTML (when fetchDetails: true and compactOutput: false)
descriptionText	Full description as plain text
descriptionMarkdown	Full description as Markdown
jobScore	0–4 quality signal
jobScoreReasons	Array of signals contributing to the score
isRepost	Whether this listing appeared in a previous run
status	Listing status from SEEK platform
scrapedAt	ISO 8601 extraction timestamp

Job score signals

jobScore is a 0–5 integer built from transparent, verifiable signals:

Signal	Score	Condition
salary_parseable	+1	Salary range is present and parseable
full_description	+1	Description is >200 characters
posted_today	+2	Posted within the last 24 hours
posted_this_week	+1	Posted within the last 7 days

Max score: 4.

---

Input examples

Basic keyword search — Hong Kong IT jobs

{
  "country": "HK",
  "searchQuery": "software engineer",
  "location": "Hong Kong",
  "maxResults": 50,
  "fetchDetails": true
}

Multi-query batch — run two searches, merged and deduplicated

{
  "country": "HK",
  "searchQuery": "[\"python developer\", \"data engineer\"]",
  "maxResults": 100,
  "fetchDetails": true
}

Post-filter — Python roles, no junior / intern titles

{
  "country": "HK",
  "searchQuery": "developer",
  "includeKeywords": "python",
  "excludeKeywords": "junior, intern, graduate",
  "maxResults": 100,
  "fetchDetails": true
}

Today's fresh listings only — Thailand

{
  "country": "TH",
  "searchQuery": "marketing manager",
  "location": "Bangkok",
  "maxAgeMinutes": 1440,
  "sortMode": "ListedDate",
  "maxResults": 50
}

Date-range research — jobs posted in a specific window

{
  "country": "HK",
  "searchQuery": "finance",
  "fromDate": "2026-06-01",
  "toDate": "2026-06-07",
  "maxResults": 200,
  "fetchDetails": true
}

Daily incremental feed — skip already-seen listings

{
  "country": "HK",
  "searchQuery": "product manager",
  "daysOld": 1,
  "skipReposts": true,
  "maxResults": 500
}

AI agent / LLM pipeline — compact high-quality output

{
  "country": "HK",
  "searchQuery": "software engineer",
  "fetchDetails": true,
  "compactOutput": true,
  "maxResults": 25
}

Sort results by jobScore descending to surface the most data-rich listings first.

---

Pricing

$0.99 per 1,000 results — you only pay for listings actually returned to your dataset. No monthly fees. No setup costs. Post-filtered results that don't reach the dataset are not charged.

Results	Cost
10	~$0.01
100	~$0.10
1,000	~$0.99
10,000	~$9.90
50,000	~$49.50

---

FAQ

Does it require a proxy? No. The SEEK API backing JobsDB responds cleanly to datacenter IPs. No proxy cost is passed on.

What markets are supported? Hong Kong (hk.jobsdb.com) and Thailand (th.jobsdb.com). HK has ~33,000 active listings; TH has ~1,200.

Can I scrape multiple keywords in one run? Yes — pass searchQuery as a JSON array: ["python", "data analyst"]. Each query runs separately and results are merged with deduplication. You pay one Actor-Start charge regardless of how many queries you run.

What is fetchDetails and when should I enable it? With fetchDetails: false (default), results come from the search index — fast, no extra requests, basic fields only. With fetchDetails: true, the actor fetches each job's detail page via the SEEK GraphQL API to add full description, salary (more accurate), email/phone extraction, and company data. Plan for roughly 4–5× more compute time.

How do post-filters work? includeKeywords, excludeKeywords, fromDate, toDate, and maxAgeMinutes are applied to the collected dataset before records are emitted. This means the actor may fetch more pages than your maxResults cap if many results are filtered out. Use daysOld as a server-side pre-filter when date range is your primary concern.

How is jobScore calculated? It's a 0–5 integer based on transparent signals: whether salary is parseable, whether a full description is present, how recently the job was posted, and whether contact details were found. See the output schema section above.

Can I use this for scheduled monitoring? Yes. Enable skipReposts: true to skip listings seen in any previous run (90-day memory). Combine with maxAgeMinutes or daysOld and a daily schedule for a continuous fresh-jobs feed.

Can I use it with the Apify API or MCP? Yes. All runs are accessible via the standard Apify API. For MCP/AI agent use, enable compactOutput: true to keep payload sizes manageable.

---

Need a custom scraper?

Unfenced Group builds Apify actors for any website — for free.

If the site you need isn't in our portfolio yet, just ask. We scope, build, and publish it at no cost to you. You only pay for results — we absorb the compute and proxy costs ourselves. Same pay-per-result pricing, same quality, same standards as every actor in this portfolio.

Get in touch: www.unfencedgroup.nl