Seek.com.au Jobs Scraper

Pricing

Rating

Developer

Actor stats

Categories

Share

Scrape jobs from Seek.com.au — titles, companies, locations, salaries, work types, arrangements, classifications, and full descriptions. HTTP-only; no login, no cookies, no proxy required.

Output (per job)

• type = job_seek
• id, url, jobLink (alias of url)
• title, company, companyLogo
• location = { city, state, postCode, country }
• locationDescription — human-readable label (e.g. "Mulgrave, Melbourne VIC")
• locationArea — Seek tracking area (e.g. "Eastern Suburbs")
• broaderLocationName — parent region (e.g. "Melbourne")
• workType — e.g. "Full Time", "Contract/Temp"
• workTypeLabels — raw enum list (e.g. ["FULL_TIME"])
• workArrangement — display text (e.g. "On-site", "Hybrid", "Remote")
• workArrangementTypes — enum list (e.g. ["ONSITE"])
• salary, salaryDetails = { min, max, currency, type }
• postedAt, postedAtLabel (e.g. "2d ago"), expiresAt
• classification, subClassification
• normalisedRoleTitle, normalisedOrganisationName
• descriptionHtml, descriptionText
• teaser — short 1-line summary
• bulletPoints (list)
• questionnaire (list of screening questions, when present)
• recruiterName, recruiterAgency, recruiterProfile
• advertiserId, advertiserName, advertiserVerified
• isPrivateAdvertiser (true when advertiser posts without a company identity)
• isLinkOut (true = external direct-apply flow)
• isVerified, isExpired, status, sourceZone
• displayType, isFeatured, isPremium, isStandout, isBranded (listing promo flags)
• applyUrl, applyLink (alias), directApplyUrl (when external)
• phoneNumber (when recruiter provides one)
• videoUrl, videoPosition (when job includes an advertiser video)
• branding = { logoUrl, coverImageUrl, thumbnailCoverImageUrl, id }
• companyId, companySlug, companyIndustry, companySize, companyWebsite
• companyDescription, companyRating, companyReviewCount, companyPerks
• companyProfile (raw nested profile object, when returned)
• emails, phoneNumbers (extracted from description text)
• scrapedAt

If a search returns zero matching jobs, a single job_seek_blocked sentinel record is emitted.

Input

How it works

1. Fetch each searchUrls entry (or build one from searchTerm + location + sortBy).
2. Parse the inline window.SEEK_REDUX_DATA JSON blob and read results.results.jobs[] — Seek ships the full job list as embedded JSON (no browser, no GraphQL auth, no pagination needed for the first page).
3. For each job, fetch https://www.seek.com.au/job/<id> to get the full description and apply link.
4. Apply client-side filters (workTypes, workArrangements, dateRange, salaryMin, salaryMax) before emitting each row.

FAQ

Do I need a proxy? No. Seek serves clean HTML to datacenter IPs.

Do I need login / cookies? No.

Why is the sentinel emitted? When the given URL / filters match no jobs, we still emit one record so downstream pipelines never see an empty output.

Are salaries always populated? No — Seek advertisers often omit salary. When it's missing the scraper simply doesn't include the salary / salaryDetails keys (no nulls).