Amazon hiring-trend tracking
Use Amazon Jobs endpoints to turn amazon hiring-trend tracking into repeatable API requests with documented inputs and JSON responses.
Search Amazon's own careers site (amazon.jobs) — by keyword, Amazon's job-category taxonomy, or country — and get a single posting by id, as structured JSON. No login, no per-category scraper to maintain.
Search Amazon's own careers site (amazon.jobs) by keyword, category, or country and get one posting by id, as structured JSON.
Endpoint families
3
Documented params
10
Examples
3
Live catalog snapshot
Active endpoints
3
Methods
GET
Required params
4
Schema refs
3
{
"platform": "Amazon Jobs",
"endpoint": "amazon-jobs-search",
"method": "GET",
"path": "/amazon-jobs/search",
"auth": "apiKey"
}Use cases
Search Amazon's own careers site (amazon.jobs) by keyword, category, or country and get one posting by id, as structured JSON.
Use Amazon Jobs endpoints to turn amazon hiring-trend tracking into repeatable API requests with documented inputs and JSON responses.
Use Amazon Jobs endpoints to turn job posting aggregation and monitoring into repeatable API requests with documented inputs and JSON responses.
Use Amazon Jobs endpoints to turn recruiting and talent-market research into repeatable API requests with documented inputs and JSON responses.
Managed execution
Every figure below is read from the live Amazon Jobs endpoint catalog — 3 endpoints, 10 documented request parameters, and 3 published response schemas — the same catalog Docs and Playground run against.
3 documented Amazon Jobs endpoints, grouped into 3 request families — Categories, Job and Search.
10 request parameters are documented across those Amazon Jobs endpoints, 4 of them required — the full input contract is public before you write any integration code.
3 of the 3 Amazon Jobs endpoints ship a recorded example response, and 3 carry a documented response schema — you can code against the real JSON before the first request.
Amazon Jobs endpoints document their error responses (400, 404, 429 and 503) alongside the success schema, so a block, a rate limit, or a missing record comes back as a typed error rather than silently empty data.
3 hosted MCP tools back the Amazon Jobs endpoints, so an agent can call the same routes with the same parameters and the same JSON contract, with no custom glue.
Coverage map
/amazon-jobs/categories
/amazon-jobs/job
/amazon-jobs/search
Endpoint catalog
/amazon-jobs/searchSearches Amazon's public careers site (amazon.jobs). Each result includes the full description and qualifications inline. `sort` accepts `relevant` (default, relevance ranking) or `recent` (newest posted first). Either `q` or `category` (or both) must be given -- `category` filters by Amazon's own job-category taxonomy and works with no text query at all. See `amazon-jobs-categories` for the full, live-verified list of accepted `category` values.
Response notes
- A valid query with no matches returns `total: 0` and an empty `jobs` array (not an error). A category can also legitimately have zero open postings at a given moment — check `amazon-jobs/categories`'s live count rather than assuming a `total: 0` response means the category value itself is wrong. - `jobs[].description`, `basic_qualifications`, and `preferred_qualifications` are HTML-stripped plain text. - `jobs[].posted_at` is best-effort parsed to RFC3339 from Amazon's free-text posted date; it is omitted if the upstream format ever drifts. - Invalid `category`, `limit` (over 100), or `sort` values return an invalid-parameter error before upstream IO. - State/city-level location filtering is not exposed — Amazon's `search.json` did not honor it reliably during research; only the country filter was confirmed working.
Sampled 14 Oct 2026: 20 items, 14 fields filled in every item, 1 not. posted_at: 2025-11-27 to 2026-10-12; country: USA (10), IND (3), AUS (2) and 4 more.
MCP tool amazon_jobs_search
/amazon-jobs/jobReturns one Amazon.jobs posting by its numeric job id (the `id` field returned by search).
Response notes
- An unknown or no-longer-listed job id returns `404`. - `description`, `basic_qualifications`, and `preferred_qualifications` are HTML-stripped plain text. - `schedule_type` and `posted_at` are only populated by the search endpoint's richer JSON payload; the detail page does not expose them, so they are omitted here.
Sampled 14 Oct 2026: 13 of 13 fields returned with a value. title: “Software Engineer”; url: amazon.jobs.
MCP tool amazon_jobs_job
/amazon-jobs/categoriesLists every value amazon-jobs/search's `category` parameter accepts, each with its current live job count. Live-queries amazon.jobs's own search category facet rather than a static list, so counts and coverage stay current.
Response notes
- `categories` is sorted by descending live `count`, ties broken alphabetically by `category`. - Every value below is currently accepted by `amazon-jobs/search`'s `category` parameter: `administrative-support`, `applied-science`, `audio-video-photography-production`, `business-intelligence`, `business-merchant-development`, `buying-planning-instock-management`, `corporate-operations`, `customer-service`, `data-science`, `database-administration`, `design`, `economics`, `editorial-writing-content-management`, `facilities-maintenance-real-estate`, `finance-accounting`, `fulfillment-associate`, `fulfillment-operations-management`, `fulfillment-warehouse-associate`, `hardware-development`, `human-resources`, `investigation-loss-prevention`, `leadership-development-training`, `legal`, `machine-learning-science`, `marketing-pr`, `medical-health-safety`, `operations-it-support-engineering`, `procurement`, `project-program-product-management-non-tech`, `project-program-product-management-technical`, `public-policy`, `public-relations`, `research-science`, `sales-advertising-account-management`, `software-development`, `solutions-architect`, `supply-chain-transportation-management`, `systems-quality-security-engineering`. - A category can report `count: 0` if amazon.jobs currently has no open postings in it; it is still a valid `category` input for `amazon-jobs/search` (it will just return an empty result page, not an error). - If amazon.jobs's own taxonomy changes (a category renamed, added, or removed upstream), this endpoint's counts reflect that immediately, but the accepted `category` slugs on `amazon-jobs/search` itself only change when the API is updated.
Sampled 14 Oct 2026: 38 items, 2 fields filled in every item. count: 0 to 3,340, median 326; category: “operations-it-support-engineering” and 37 other values.
MCP tool amazon_jobs_categories
Related APIs
Business & Market Intelligence
Turn public GitHub ecosystem signals into structured data for developer-market research, open-source intelligence, recruiting, and AI-agent workflows.
Business & Market Intelligence
Collect public TrustMRR startup marketplace and leaderboard data as normalized JSON for startup research, SaaS discovery, and market intelligence workflows.
Business & Market Intelligence
Add website and competitive intelligence signals from SimilarWeb to research pipelines.
Related dataset
Dataset · queryable JSON
Amazon's open postings are also queryable in bulk alongside 18 other providers — filter by department, location, employment type, or remote flag over one REST call.
Explore the dataset →How to scrape Amazon Jobs
amazon.jobs runs its own search API, separate from any shared ATS — a plain scraper has to reverse-engineer Amazon's own category taxonomy and pagination. Crawlora's Amazon Jobs endpoints return normalized postings directly, no per-category scraper to maintain. A pull on 2026-10-27 showed 38 job categories led by operations-it-support-engineering with 3,105 openings, and 416 results for data engineer in the USA at 20 per page.
/amazon-jobs/search takes q, category, country, page, limit and sort (relevant or a recency order). Each result includes the full description and qualifications inline, so one call is enough to rank or classify postings.
/amazon-jobs/categories lists every value the search category parameter accepts, each with its current live job count. Pass a returned slug rather than a typed label, and refresh the list because the counts move.
/amazon-jobs/job takes the numeric id field returned by search and returns that single posting.
categories returned 38 slugs with live counts (operations-it-support-engineering showed 3,105), fetched the day before the search pull. Use the slug as the category filter.
search for data engineer in USA returned total 416 and 20 jobs per page. Each job has an id (10537982 for Data Engineer, SPTC), location split into city, state and country, a category, a schedule type and a posted_at date such as 2026-09-11.
Agent skills
Install a Crawlora agent skill and your agent can pull Amazon Jobs data as JSON through the documented API, without writing a scraper. These skills cover Amazon Jobs:
Researches Amazon.jobs postings via the Crawlora API — search Amazon's public careers site by keyword and/or job category, then pull any single…
npx skills add github.com/Crawlora-org/crawlora-skills --skill amazon-jobs-researchResearches job postings, hiring signals, and freelance gigs via the Crawlora API — Indeed, Google/Amazon/Apple/Meta/Tesla careers sites, any…
npx skills add github.com/Crawlora-org/crawlora-skills --skill job-market-researchFAQ
Send a keyword, category, or country to Crawlora's Amazon Jobs search endpoint and get normalized postings — title, location, category, apply URL — without reverse-engineering amazon.jobs's own category taxonomy or pagination. Collect only public data and respect Amazon's terms.
Amazon's own taxonomy — software-development, data-science, business-intelligence-data-engineering, fulfillment-center-warehouse-associate, and dozens more — passed as the category parameter; either category or a free-text q is required.
Yes — pass the numeric Amazon job id to the single-posting endpoint for the full listing.
20 per page. On 2026-10-27 the query data engineer in the USA reported 416 in total.