No job links match, but the board is not actually empty

Template drift, a truncated body, or an interstitial can all produce zero matches. Only treat the board as empty when the explicit gnewtonNoActiveJobs sentinel is present; otherwise retry rather than persisting a zero-job snapshot.

Board renders as JavaScript arrays instead of anchors

Tenants on the sort-by-department view emit jobName[dep][i] and jobId[dep][i] arrays rather than tags. Try anchor parsing (Template A) first, then fall back to pairing the JS arrays by department and index (Template B).

Query strings are sometimes HTML-escaped (& vs &)

Some tenants escape the href query string and some don't. HTML-unescape each href before parsing and match both & and & separators so clientId and id extraction is reliable.

Detail page marked 'this job is no longer active'

The description table is missing on inactive landings. Detect the inactive sentinel and skip the record instead of storing empty or stale HTML as a valid job.

Duplicate or cross-client anchors on the page

Dedupe listings by their native hex job id and skip any anchor whose clientId does not match the tenant you are scraping, so shared or template rows don't create phantom jobs.

Requests blocked or rate limited (403 / 429)

Paycor throttles aggressive scraping. Space requests ~250ms apart, cap concurrent detail fetches at ~3, and retry transient network/5xx failures a couple of times with a short backoff.

Paycor Jobs API.

Pull structured job titles, locations, and full descriptions from company career boards hosted on Paycor's recruiting (gnewton) platform at recruitingbypaycor.com.

Get API access

Data fields

Full HTML job descriptions
Job titles
Office & remote locations
Department & team grouping
Native hex job IDs
Openings count

Use cases

01Career board aggregation
02SMB hiring signals
03Local job market tracking
04Recruiting data pipelines

Validate the URL and extract the clientId

Every Paycor board lives on recruitingbypaycor.com and is keyed by a hexadecimal clientId (16+ hex chars) in the query string. Confirm the host and clientId before fetching, then build the canonical CareerHome.action listings URL.

from urllib.parse import urlparse, parse_qs
import re

url = "https://recruitingbypaycor.com/career/CareerHome.action?clientId=8a7883c67239c84401727af009b53231"

parsed = urlparse(url)
assert parsed.netloc == "recruitingbypaycor.com", "Not a Paycor recruiting host"

client_id = (parse_qs(parsed.query).get("clientId") or [""])[0]
assert re.fullmatch(r"[0-9a-fA-F]{16,}", client_id), "Missing or malformed clientId"

listings_url = f"https://recruitingbypaycor.com/career/CareerHome.action?clientId={client_id}"
print(listings_url)

Fetch the career board HTML

Paycor serves the board as plain server-rendered HTML — there is no JSON endpoint (format=json still returns text/html and /career/api/jobs returns 404). Request the page with an HTML Accept header and a short timeout.

import requests

resp = requests.get(
    listings_url,
    headers={"Accept": "text/html,application/xhtml+xml"},
    timeout=15,
)
resp.raise_for_status()
page_html = resp.text

Parse Template A (server-rendered anchors)

Most tenants render a sort-by-location board with anchor tags pointing at JobIntroduction.action. Extract each job's clientId and id, skip cross-client links and duplicates, and build the canonical detail URL. Location group headers appear in div.gnewtonCareerGroupHeaderClass elements that precede each run of rows.

from bs4 import BeautifulSoup
from urllib.parse import urlparse, parse_qs
import html

soup = BeautifulSoup(page_html, "html.parser")

listings = []
seen = set()
for a in soup.select('a[href*="JobIntroduction.action"]'):
    href = html.unescape(a.get("href", ""))
    qs = parse_qs(urlparse(href).query)
    href_client = (qs.get("clientId") or [""])[0]
    job_id = (qs.get("id") or [""])[0]

    # Skip cross-client links and repeated job IDs.
    if href_client.lower() != client_id.lower() or not job_id or job_id in seen:
        continue
    seen.add(job_id)

    listings.append({
        "external_id": f"{client_id}:{job_id}",
        "job_id": job_id,
        "title": a.get_text(strip=True),
        "detail_url": (
            "https://recruitingbypaycor.com/career/JobIntroduction.action"
            f"?clientId={client_id}&id={job_id}&source=&lang=en"
        ),
    })

print(f"Template A found {len(listings)} jobs")

Fall back to Template B (JavaScript arrays)

Tenants using the sort-by-department view emit listings as JavaScript arrays (jobName[dep][i] and jobId[dep][i]) instead of anchors. If Template A yields nothing, pair the two arrays by department and index to recover titles and job IDs.

import re

if not listings:
    names = {}
    for dep, idx, title in re.findall(
        r'jobName\["([0-9a-fA-F]{16,})"\]\[(\d+)\s*-\s*1\]\s*=\s*"([^"]*)"',
        page_html,
    ):
        names[(dep, idx)] = title

    for dep, idx, job_id in re.findall(
        r'jobId\["([0-9a-fA-F]{16,})"\]\[(\d+)\s*-\s*1\]\s*=\s*"([0-9a-fA-F]{16,})"',
        page_html,
    ):
        if any(l["job_id"] == job_id for l in listings):
            continue
        listings.append({
            "external_id": f"{client_id}:{job_id}",
            "job_id": job_id,
            "title": names.get((dep, idx), ""),
            "detail_url": (
                "https://recruitingbypaycor.com/career/JobIntroduction.action"
                f"?clientId={client_id}&id={job_id}&source=&lang=en"
            ),
        })

    print(f"Template B found {len(listings)} jobs")

Fetch full job details

The board only carries titles and locations, so request each JobIntroduction.action detail page for the full description. Paycor renders labeled cells with stable gnewton ids: #gnewtonJobPosition (title), #gnewtonJobDescriptionText (description HTML), #gnewtonJobLocationInfo (one or more locations), plus #gnewtonJobID and #gnewtonJobOpening.

import re
import requests
from bs4 import BeautifulSoup

def fetch_detail(detail_url: str) -> dict | None:
    resp = requests.get(
        detail_url,
        headers={"Accept": "text/html,application/xhtml+xml"},
        timeout=15,
    )
    resp.raise_for_status()

    # Skip stale pages that no longer describe a live job.
    if re.search(r"this\s+job\s+is\s+no\s+longer\s+active", resp.text, re.I):
        return None

    soup = BeautifulSoup(resp.text, "html.parser")

    def cell(selector: str, strip_label: bool = False) -> str | None:
        el = soup.select_one(selector)
        if el is None:
            return None
        text = el.get_text(" ", strip=True)
        if strip_label and ":" in text:
            text = text.split(":", 1)[1].strip()
        return text or None

    desc_el = soup.select_one("#gnewtonJobDescriptionText")
    return {
        "title": cell("#gnewtonJobPosition", strip_label=True),
        "description_html": desc_el.decode_contents().strip() if desc_el else None,
        "locations": [el.get_text(" ", strip=True) for el in soup.select("#gnewtonJobLocationInfo")],
        "client_job_id": cell("#gnewtonJobID", strip_label=True),
        "openings": cell("#gnewtonJobOpening", strip_label=True),
    }

for job in listings[:3]:
    print(fetch_detail(job["detail_url"]))

Distinguish an empty board from a parse failure

Never record zero jobs just because no links matched — that can mean template drift, a truncated body, or an interstitial. Only treat a board as genuinely empty when Paycor renders its explicit gnewtonNoActiveJobs sentinel; otherwise retry rather than persisting an empty snapshot.

import re

if not listings:
    if re.search(r'id\s*=\s*["\']gnewtonNoActiveJobs["\']', page_html, re.I):
        print("Board explicitly reports no active jobs — safe to record as empty")
    else:
        raise RuntimeError(
            "No jobs parsed and no empty-state sentinel found — "
            "treat as incomplete and retry, do not record zero jobs"
        )

Common issues

highNo job links match, but the board is not actually empty: Template drift, a truncated body, or an interstitial can all produce zero matches. Only treat the board as empty when the explicit gnewtonNoActiveJobs sentinel is present; otherwise retry rather than persisting a zero-job snapshot.
mediumBoard renders as JavaScript arrays instead of anchors: Tenants on the sort-by-department view emit jobName[dep][i] and jobId[dep][i] arrays rather than <a> tags. Try anchor parsing (Template A) first, then fall back to pairing the JS arrays by department and index (Template B).
mediumQuery strings are sometimes HTML-escaped (& vs &): Some tenants escape the href query string and some don't. HTML-unescape each href before parsing and match both & and & separators so clientId and id extraction is reliable.
mediumDetail page marked 'this job is no longer active': The description table is missing on inactive landings. Detect the inactive sentinel and skip the record instead of storing empty or stale HTML as a valid job.
lowDuplicate or cross-client anchors on the page: Dedupe listings by their native hex job id and skip any anchor whose clientId does not match the tenant you are scraping, so shared or template rows don't create phantom jobs.
mediumRequests blocked or rate limited (403 / 429): Paycor throttles aggressive scraping. Space requests ~250ms apart, cap concurrent detail fetches at ~3, and retry transient network/5xx failures a couple of times with a short backoff.

Best practices

1Validate the host is recruitingbypaycor.com and the clientId is 16+ hex chars before fetching
2Try Template A anchors first, then fall back to the Template B JavaScript arrays
3HTML-unescape hrefs and match both & and & separators when extracting IDs
4Only record an empty board when the gnewtonNoActiveJobs sentinel is present
5Dedupe listings by native hex job id and drop cross-client anchors
6Fetch detail pages for full descriptions, locations, and openings count; keep ~250ms between requests

Or skip the complexity

One endpoint. All Paycor jobs. No scraping, no sessions, no maintenance.

Get API access

curl "https://connect.jobo.world/api/jobs?sources=paycor" \
  -H "X-Api-Key: YOUR_KEY"

Ready to integrate

Access Paycor
job data today.

One API call. Structured data. No scraping infrastructure to build or maintain — start with the $5 free starting balance and scale as you grow.

Get API access View all platforms

99.9%API uptime

<200msAvg response

50M+Jobs processed

Paycor Jobs API.

What's in every response.

Data fields

Use cases

How to scrape Paycor.

Validate the URL and extract the clientId

Fetch the career board HTML

Parse Template A (server-rendered anchors)

Fall back to Template B (JavaScript arrays)

Fetch full job details

Distinguish an empty board from a parse failure

Try the API.

Access Paycor
job data today.

Paycor Jobs API.

Data fields

Use cases

Validate the URL and extract the clientId

Fetch the career board HTML

Parse Template A (server-rendered anchors)

Fall back to Template B (JavaScript arrays)

Fetch full job details

Distinguish an empty board from a parse failure

Access Paycorjob data today.

Access Paycor
job data today.