Apple Jobs Jobs API.

import requests

session = requests.Session()

boot = session.get(
    "https://jobs.apple.com/api/v1/CSRFToken",
    headers={"Accept": "application/json"},
)
boot.raise_for_status()

# The token comes back on a response header; the cookies stay on the session.
csrf_token = boot.headers["X-Apple-CSRF-Token"]

def fetch_page(session, csrf_token, page):
    resp = session.post(
        "https://jobs.apple.com/api/v1/search",
        headers={
            "Accept": "application/json",
            "Content-Type": "application/json",
            "X-Apple-CSRF-Token": csrf_token,
            "Origin": "https://jobs.apple.com",
            "Referer": "https://jobs.apple.com/en-gb/search?location=united-kingdom-GBR",
        },
        json={
            "query": "",
            "filters": {"locations": ["postLocation-GBR"]},
            "page": page,
            "locale": "en-gb",
            "sort": "",
            "format": {"longDate": "MMMM D, YYYY", "mediumDate": "MMM D, YYYY"},
        },
    )
    resp.raise_for_status()
    return resp.json()["res"]

PAGE_SIZE = 20

def fetch_all(session, csrf_token):
    seen, jobs, page = set(), [], 1
    while True:
        res = fetch_page(session, csrf_token, page)
        rows = res.get("searchResults", [])
        total = res.get("totalRecords", 0)
        for row in rows:
            position_id = row.get("positionId")
            if not position_id or position_id in seen:
                continue
            seen.add(position_id)
            jobs.append(row)
        if not rows or page * PAGE_SIZE >= total:
            break
        page += 1
    return jobs

from urllib.parse import quote

def to_record(row):
    position_id = row["positionId"]
    slug = row.get("transformedPostingTitle") or "job"
    team = row.get("team") or {}
    return {
        "id": position_id,
        "title": row.get("postingTitle"),
        "team": team.get("teamName"),
        "team_code": team.get("teamCode"),
        "locations": [loc.get("name") for loc in row.get("locations", [])],
        "posted_at": row.get("postDateInGMT") or row.get("postingDate"),
        "url": f"https://jobs.apple.com/en-gb/details/{quote(position_id)}/{quote(slug)}",
        "apply_url": f"https://jobs.apple.com/app/en-gb/apply/{quote(position_id)}",
    }

import json, re

HYDRATION = re.compile(
    r'window\.__staticRouterHydrationData\s*=\s*JSON\.parse\('
    r'(?P<json>"(?:\\.|[^"\\])*")\);'
)

def fetch_details(session, detail_url):
    html = session.get(detail_url, headers={"Accept": "text/html"}).text
    match = HYDRATION.search(html)
    if not match:
        raise ValueError("Apple detail page omitted hydration data")

    # The payload is a JSON string literal wrapping the real JSON document.
    payload = json.loads(json.loads(match.group("json")))
    job = payload["loaderData"]["jobDetails"]["jobsData"]
    return {
        "title": job.get("postingTitle"),
        "employment_type": job.get("employmentType"),
        "summary": job.get("jobSummary"),
        "responsibilities": job.get("responsibilities"),
        "minimum_qualifications": job.get("minimumQualifications"),
        "preferred_qualifications": job.get("preferredQualifications"),
    }