New jobs

Active jobs we first collected in a time window, as a JSON array. This is the feed to poll when you keep your own copy of the catalogue: call it on a fixed schedule and every new job reaches you once.

GET/v1/active-ats

Time frames

Value	Use	Window
1h	Hourly polling	The last full UTC hour. Called at 09:15 it returns jobs collected 08:00–09:00. Each hour becomes available two minutes after it ends.
24h	Daily polling (default)	The 24 hours up to the last full UTC hour. Call it in the same hour every day.
7d	Short backfill	Rolling 7 days up to now.
6m	Full backfill	Every active job collected in the last 183 days, which is essentially the whole catalogue. Page it with `cursor`.

Why windows end on the hour

A window that ends at a fixed hour boundary returns the same jobs whenever in the next hour you call it, so hourly or daily polling never skips or repeats a job. For the freshest data, poll 1h a few minutes after each hour, or use the changes feed, which is seconds behind collection.

Parameters

Pagination and response

Parameter	Type	Description
time_frame	string	`1h`, `24h`, `7d` or `6m`. Default `24h`.
limit	integer	Jobs per page, 1–1000. Default `100`.
offset	integer	Jobs to skip. Pages are ordered by `date_posted`, newest first (jobs without a posted date use `date_created`).
cursor	integer	Pass the last `id` of the previous page. Orders by `id` ascending and overrides `offset`. Use it for `6m` backfills.
description_format	string	`text` or `html`. Omit to leave descriptions out, which keeps pages small.
include_source_fields	boolean	`true` adds `source_fields`: department, pay tiers, qualifications and other facts exactly as the employer published them, with where each was found.
include_basic_organization_details, include_beta	boolean	Accepted for compatibility; they add no fields.

Search

Parameter	Type	Description
title	string	Search the title. `nurse manager` needs both words, `"nurse manager"` is a phrase, `nurse OR manager` either, `-assistant` excludes, `engineer*` is a prefix.
description	string	Same syntax, over the title, employer name and description together.
location	string	Full place names, `City, Region, Country`, combined with `OR`. See Location search.
title_advanced, description_advanced, location_advanced, organization_advanced	string	Boolean expressions with `&`, `\|`, `!`, `<->`, parentheses and `:*`. They take precedence over the plain parameter. See Advanced search.

Dates

Parameter	Type	Description
date_posted_gte, date_posted_lt	ISO 8601	When the employer says the job was posted. A time without a zone is UTC.
date_created_gte, date_created_lt	ISO 8601	When we first collected the job. Use these to replay a missed window exactly. See Time fields.

Filters

Parameter	Type	Description
country	list	`CA`, `US` or both (the default).
organization, exclude_organization	list	Exact, case-sensitive employer names, comma-separated. Quote a name that contains a comma: `"Acme, Inc."`. The organization list has every name.
source, exclude_source	list	Source systems, e.g. `workday,icims`. See Sources.
has_salary	boolean	`true`: the employer published pay. `false`: it did not.
remote	boolean	`true`: marked remote by the employer. `false`: not remote.
has_no_location	boolean	`true`: jobs without a resolved location.
id	list	Only these job ids.

Unknown parameters are rejected with a suggestion (timeframe → time_frame) instead of being ignored, so a typo never silently widens your results.

Example

bash

curl --get "https://api.thejobsapi.com/v1/active-ats" \
  -H "Authorization: Bearer $THEJOBSAPI_KEY" \
  --data-urlencode 'time_frame=1h' \
  --data-urlencode 'title="registered nurse" OR rn' \
  --data-urlencode 'location="Ontario, Canada" OR "New York, United States"' \
  --data-urlencode 'description_format=text'

Pagination

For 1h, 24h and 7d, page with offset: request limit=1000&offset=0, then offset=1000, and so on until a page has fewer than limit jobs.

For 6m, use cursor: start with cursor=0 and pass the last id of each page. Cursor pages are ordered by id and stay fast however deep you go.

Backfill with cursor

let cursor = 0;
for (;;) {
  const url = `https://api.thejobsapi.com/v1/active-ats?time_frame=6m&limit=1000&cursor=${cursor}&description_format=html`;
  const response = await fetch(url, { headers: { Authorization: `Bearer ${process.env.THEJOBSAPI_KEY}` } });
  if (!response.ok) throw new Error((await response.json()).detail);
  const jobs = await response.json();
  await saveJobs(jobs);
  if (jobs.length < 1000) break;
  cursor = jobs[jobs.length - 1].id;
}

Response

An array of jobs. Fields we don't collect are null; see Job fields for every field and where it comes from.

json

[
  {
    "id": 4183022,
    "title": "Registered Nurse, Emergency",
    "organization": "Example Health",
    "organization_url": null,
    "organization_logo": null,
    "date_posted": "2026-10-12T00:00:00Z",
    "date_created": "2026-10-12T14:31:07Z",
    "date_valid_through": "2026-11-01T00:00:00Z",
    "locations": [{
      "@type": "Place",
      "address": { "@type": "PostalAddress", "addressLocality": "Toronto", "addressRegion": "ON", "addressCountry": "CA" },
      "latitude": 43.70011, "longitude": -79.4163
    }],
    "locations_alt": ["Toronto, ON"],
    "locations_derived": ["Toronto, Ontario, Canada"],
    "location_type": null,
    "location_requirements": null,
    "salary": {
      "@type": "MonetaryAmount", "currency": "CAD",
      "value": { "@type": "QuantitativeValue", "minValue": 42.5, "maxValue": 51.0, "unitText": "HOUR" }
    },
    "employment_type": ["FULL_TIME"],
    "url": "https://example.wd3.myworkdayjobs.com/careers/job/Toronto/Registered-Nurse_R12345",
    "source": "workday",
    "source_type": "ats",
    "source_domain": "example.wd3.myworkdayjobs.com",
    "source_slug": "example.wd3.myworkdayjobs.com/careers",
    "domain_derived": null,
    "cities_derived": ["Toronto"],
    "counties_derived": null,
    "regions_derived": ["Ontario"],
    "countries_derived": ["Canada"],
    "timezones_derived": null,
    "lats_derived": [43.70011],
    "lngs_derived": [-79.4163],
    "country": "CA",
    "apply_url": null,
    "work_arrangement": "onsite"
  }
]

Usage

Each job returned counts once toward your plan's job allowance; each call counts as one request. See Errors, limits & usage.