iRecruit US Jobs API.

import re
from urllib.parse import urlparse, parse_qs

HOST = "www.irecruit-us.com"
ORG_ID = re.compile(r"^I\d{8}$")
REQUEST_ID = re.compile(r"^[A-Za-z0-9]{10,64}$")

def parse_board(url: str) -> tuple[str, str] | None:
    parsed = urlparse(url)
    if parsed.netloc.lower() != HOST or parsed.path.lower() != "/index.php":
        return None
    query = parse_qs(parsed.query)
    org = (query.get("OrgID") or [""])[0].upper()
    if not ORG_ID.match(org):
        return None

    # Category / level selections are part of the board identity, not noise.
    suffix = ""
    if "olnew" in query and "slnew" in query:
        suffix = f"&olnew={query['olnew'][0]}&slnew={query['slnew'][0]}"
    elif "level" in query:
        suffix = f"&level={query['level'][0]}"
        if "menulist" in query:
            suffix += f"&menulist={query['menulist'][0]}"
    return org, f"https://{HOST}/index.php?OrgID={org}&navpg=listings{suffix}"

org, board_url = parse_board("https://www.irecruit-us.com/index.php?OrgID=I20100401&navpg=listings")
print(board_url)

import requests
from bs4 import BeautifulSoup

def fetch_board(session: requests.Session, org: str, board_url: str) -> BeautifulSoup:
    response = session.get(
        board_url,
        headers={"Accept": "text/html,application/xhtml+xml"},
        timeout=30,
    )
    response.raise_for_status()

    # iRecruit prints its own organization id into the page script.
    if not re.search(rf"var\s+OrgID\s*=\s*['\"]{org}['\"]", response.text, re.IGNORECASE):
        raise RuntimeError("iRecruit page omitted its first-party organization proof")
    if parse_board(response.url) != (org, board_url):
        raise RuntimeError("iRecruit board redirected outside the requested organization")
    return BeautifulSoup(response.text, "html.parser")

session = requests.Session()
soup = fetch_board(session, org, board_url)

from urllib.parse import urljoin

def collect_listings(soup: BeautifulSoup, org: str, board_url: str) -> list[dict]:
    listings, seen = [], set()
    for anchor in soup.select("a[href*='RequestID=']"):
        target = urljoin(board_url, anchor["href"])
        parsed = urlparse(target)
        query = parse_qs(parsed.query)
        request_id = (query.get("RequestID") or [""])[0]
        if (query.get("OrgID") or [""])[0].upper() != org:
            continue
        if not REQUEST_ID.match(request_id) or request_id in seen:
            continue
        bold = anchor.find("b")
        title = " ".join((bold or anchor).get_text().split())
        if not title:
            continue
        seen.add(request_id)
        listings.append({"id": request_id, "title": title, "listing_url": target})
    return listings

listings = collect_listings(soup, org, board_url)
print(f"{len(listings)} vacancies on {org}")

def fetch_detail(session: requests.Session, org: str, listing: dict) -> dict | None:
    response = session.get(listing["listing_url"], timeout=30)
    if response.status_code in (404, 410):
        return None  # canonical removal
    response.raise_for_status()
    soup = BeautifulSoup(response.text, "html.parser")

    def has_hidden(name: str, value: str) -> bool:
        return any(
            (field.get("value") or "").upper() == value.upper()
            for field in soup.select(f"input[name='{name}']")
        )

    # A live request keeps its native application form. Without it the server has
    # returned the organization surface for a request that no longer exists.
    if not has_hidden("OrgID", org) or not has_hidden("RequestID", listing["id"]):
        return None

    container = soup.select_one("#reqbuttons")
    container = container.parent if container else soup.select_one("#page-wrapper")
    body = container.decode_contents() if container else ""
    body = body.split('<div id="reqbuttons"', 1)[0]  # drop the apply controls

    location = re.search(r"<b>\s*Location:\s*</b>\s*([^<]+)", body, re.IGNORECASE)
    meta_title = soup.select_one("meta[property='og:title']")
    return {
        **listing,
        "title": (meta_title.get("content") if meta_title else None) or listing["title"],
        "description_html": body.strip() or None,
        "location": " ".join(location.group(1).split()) if location else None,
    }

for listing in listings[:3]:
    job = fetch_detail(session, org, listing)
    print(job["title"], "-", job["location"] if job else "no longer posted")