GulfTalent Jobs Scraper — stale listings flagged
Pricing
$5.00 / 1,000 job scrapeds
GulfTalent Jobs Scraper — stale listings flagged
Scrape GulfTalent listings and get what the result count hides: exact posting dates instead of vague labels, how many postings are stale, and how many are the same role posted more than once. The headline count is almost never the number of real openings. Unofficial; not affiliated with GulfTalent.
Pricing
$5.00 / 1,000 job scrapeds
Rating
0.0
(0)
Developer
Entrogix Works
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
6 days ago
Last modified
Categories
Share
Scrape GulfTalent listings and get what the result count hides: exact posting dates, how many listings are stale, and how many are the same role posted more than once.
Unofficial. Not affiliated with, endorsed by, or connected to GulfTalent.
Why this one
"237 results" is not 237 openings. Some of those postings are months old, and some are the same job listed twice by the same employer. If you're sizing a market, tracking a competitor's hiring, or deciding where to apply, the headline number is the wrong number.
| Field | What it answers |
|---|---|
postedDate / daysOld | How old is this, exactly |
isStale | Older than 30 days |
duplicateCount | How many times this same role appears |
isDuplicate | Is this a repeat of an earlier listing |
The GULFTALENT_SUMMARY record turns that into the one line people actually want:
listings shown vs. distinct roles, plus the duplicate rate and the median age.
Duplicates are flagged, never deleted
Every row is returned. If repeats were dropped, you could no longer reconcile the output against GulfTalent's own count, and you'd have no way to check the grouping was right. Grouping is by company + normalised title within a single run — which is why the input takes one country at a time.
The year is inferred, and it says so
GulfTalent prints the day and month only ("12 Jul"). The year comes from the run date,
with one rule: a date that would land in the future belongs to last year. Without
that, a December posting read in August becomes four months in the future and every
staleness figure inverts. postedText ships alongside so you can audit the parse.
Dates that cannot be read become null, never today. dateParsedRate in the summary
tells you what share parsed cleanly.
Company can be null
GulfTalent puts the job title and the company in the same table cell. The company is
recovered by removing the title from the front of that cell — and when that doesn't come
apart cleanly, the field is null rather than a mangled string.
What you get
Per listing: job ID, title, company, location, posted date, days old, stale flag, duplicate count and repeat flag, job URL, company URL.
A GULFTALENT_SUMMARY record per run: listings shown, distinct roles, duplicate
count and rate, median days old, stale count, what share of dates parsed, the oldest
posting date, and the top companies and locations.
Coverage
uae, saudi-arabia, qatar, kuwait, bahrain, oman.
Reliability
Success rate was measured before this Actor was written: 5/5 through the same browser path used in production. It is monitored twice daily against a live page.
Proxy settings were measured too, not assumed: no proxy 6/6, datacenter 6/6. The default is no proxy, which is the cheapest option available. Turn it on only if you hit blocks.
Input
{"queries": ["data analyst", "software engineer"],"country": "uae","maxItemsPerQuery": 50}
Plain search terms become GulfTalent list URLs for the chosen country. Full list URLs
work too, in which case country is ignored.
Billing
Pay per event: one job-scraped event per listing actually delivered to your dataset.
Duplicate rows (the same listing fetched twice across pages) are removed before charging.
Repeat postings are real listings and are charged — they're part of what you asked for.
A run that returns nothing costs nothing.


