By the end of this tutorial you will have a Python script that pulls job listings from karriere.at, the Austrian job board, into structured rows: title, employer, location, salary text and the home-office flag.
The karriere.at Jobs Scraper (reventadata/karriere-scraper) is the Apify Actor built for this site. You give it one or more keywords, and it returns one row per job the search lists. This guide runs it from Python with the Apify API client.
The rows you see below come from a real run on 17 November 2026, with the input shown in the script. Nothing in them was edited. Only null fields are left out of the sample row.
Prerequisites
You need three things:
- Python 3.11 or newer, because the Apify client for Python 3.x requires it.
- An Apify account, which gives you the API token.
- Your Apify API token, exported as an environment variable.
Install the client and export the token:
pip install -U "apify-client~=3.0"
export APIFY_API_TOKEN="your_apify_token_here"
No scraping code is needed on your side. The Actor reads the public pages on karriere.at and returns the rows.
What you'll achieve
You will search karriere.at for the keyword python, take up to 10 jobs and print one line per job.
The limit of 10 comes from the input, not from the script. maxItems caps the rows saved across all keywords, and maxPagesPerQuery caps the result pages read for each keyword. In this run both were set, so the run stopped at 10 rows.
In this run 9 of the 10 titles contain the word Python. The tenth, "Senior Integration & AI Automation Engineer (all genders)", has Python in its summary text.
Use the same script for a different keyword, a city, or a home-office-only search. Only the input changes.
Step 1: Find the Actor
The Actor's Apify name is reventadata/karriere-scraper. The Python client calls it by that name, so you do not need to open the Store page to run the script.
Step 2: Choose the search
These are the input fields from the Actor's input form that the script uses or that you may add:
| Field | Type | What it does |
|---|---|---|
searchQueries |
array of strings | Keywords, one search each. An empty list searches by location alone. |
location |
string | A karriere.at location slug in lower case, such as wien or graz. Empty means all of Austria. |
homeOffice |
boolean | When true, returns only jobs the site marks as home office. The default is false. |
fetchDetails |
boolean | Opens each job's page to add the full description, the exact posting date, structured salary and the employer website. Slower. The default is false. |
maxItems |
integer | The maximum number of jobs saved across all keywords. If you leave it out, the Actor saves up to 100. |
maxPagesPerQuery |
integer | Result pages read for each keyword, from 1 to 500. |
A location must be a karriere.at slug: lower-case letters, digits and hyphens only. Any other value makes the run end at once as FAILED, with a message that names the input to fix.
Step 3: Run the Actor and wait
Save this as karriere_jobs.py:
import os
from apify_client import ApifyClient
# 1. Initialise the client with your token from the environment
client = ApifyClient(os.environ["APIFY_API_TOKEN"])
# 2. Describe the search: the keyword "python", 10 jobs at most, one page of results
run_input = {
"searchQueries": ["python"],
"maxItems": 10,
"maxPagesPerQuery": 1,
"homeOffice": False,
"fetchDetails": False,
}
# 3. Start the Actor and wait for it to finish
run = client.actor("reventadata/karriere-scraper").call(run_input=run_input)
if run is None or run.status != "SUCCEEDED":
status = run.status if run else "unknown"
raise SystemExit(f"Run did not succeed (status: {status}).")
# 4. Read the rows from the run's default dataset
items = client.dataset(run.default_dataset_id).list_items().items
print(f"Retrieved {len(items)} jobs.\n")
# 5. Print one line per job: title, employer, first location, salary text
for item in items:
print(f"{item['title']} | {item['companyName']} | {item['location']} | {item['salaryText']}")
Run it with python3 karriere_jobs.py. The call waits until the run finishes. In this run the Actor finished in 5.4 seconds.
Step 4: Read the output
The script prints this for the run above:
Retrieved 10 jobs.
Senior Integration & AI Automation Engineer (all genders) | IMS Nanofabrication GmbH | Brunn am Gebirge | ab 70.000 € jährlich
Lead Software Developer Python/Java (w/m/d) | APG - Austrian Power Grid AG | Wien | ab 5.213,07 € monatlich
Testing, Test Automation & Quality Assurance mit Python (w/m/d) | jonnect | Wien | 3.500 € – 5.000 € monatlich
Odoo-Entwickler (m/w/d) – Python & ERP | Grüne Erde GmbH | Scharnstein | ab 3.300 € monatlich
Software Developer - Odoo / Python (m/w/d) | BLOCH3 Projektentwicklung GmbH | Wien | ab 55.000 € jährlich
Applikationsingenieur:in / Python-Developer:in – Video- & Sicherheitssysteme | Siemens Personaldienstleistungen GmbH | Graz | ab 3.449,08 € monatlich
Softwareentwickler*in C#/Python (m/w/d) | IVM Technical Consultants | Wels | ab 4.400 € monatlich
Senior Python & Java Developer (w/m/d) | Bundesrechenzentrum GmbH | Wien | ab 69.011,60 € jährlich
Software Entwickler mit Fokus Python (w/m/x) | epunkt GmbH | Wien | ab 3.400 € monatlich
Software Engineer Python mit Schwerpunkt Knowledge Graphs & Schema-Automatisierung (w/m/div.) | Bosch-Gruppe Österreich | Linz | ab 53.000 € jährlich
Each dataset row carries more fields than the script prints. This is one row, copied from the run's dataset. Null fields are left out, and so is searchLocation, which is null on all 10 rows because this run set no location:
{
"id": "7881766",
"url": "https://www.karriere.at/jobs/7881766",
"title": "Softwareentwickler*in C#/Python (m/w/d)",
"companyName": "IVM Technical Consultants",
"companyId": "1740",
"companyUrl": "https://www.karriere.at/f/ivm-technical-consultants",
"companyLogoUrl": "https://kcdn.at/company/1740/3745682/ivm_karriere_at.companysquare.png",
"companySize": "101 - 500",
"companyMainLocation": "Wien Vösendorf",
"location": "Wels",
"locations": [
"Wels"
],
"isHomeOffice": false,
"employmentType": "Vollzeit",
"salaryText": "ab 4.400 € monatlich",
"summary": "Entwicklung und Weiterentwicklung von Softwarelösungen im Bereich KI, Machine Learning und datengetriebenen Systemen. Erfahrung in C# und/oder Python ist erforderlich. Sie arbeiten an abwechslungsreichen Projekten mit modernen KI-Themen.",
"postedText": "vor 5 Tagen veröffentlicht",
"isPromoted": false,
"isSmartApply": false,
"searchKeyword": "python",
"scrapedAt": "2026-11-17T21:59:15+00:00"
}
Understanding your results
Each row has the same fields. Here is what they hold in this run:
-
locationandlocations.locationis the first work location.locationslists all of them. One row lists two: Wien and Graz. -
isHomeOffice. True on 8 of the 10 rows, false on 2. -
salaryText. The salary as the listing writes it. "ab" means "from". Nine rows give a starting figure and one gives a range (3.500 € – 5.000 € monatlich). Six rows give a monthly amount and four a yearly one. -
postedText. The posting date as the listing shows it. Eight rows show a relative text such as "vor 5 Tagen veröffentlicht" (published 5 days ago). One shows "Heute veröffentlicht" (published today), and one shows the absolute date 8.11.2026. -
companySize. The employer's size bucket as listed. Four rows show "501 - 2500". -
isPromotedandisSmartApply. The listing's own flags.isPromotedis true on 2 rows andisSmartApplyon 3.
Fields that are empty in this run:
-
description,postedAt,salaryMin,salaryMax,salaryValue,salaryCurrency,salaryPeriod,employmentTypeCodes,addressRegion,employerWebsiteanddirectApplyare null on all 10 rows. This run setfetchDetailsto false. -
searchLocationis null on all 10 rows. This run set no location. -
summaryis null on one row (id 10038942).
The 10 rows have 10 different job ids. The Actor skips an id it has already saved, so one job never appears twice in a run.
Pro tips and advanced usage
-
Narrow by city. Set
locationto a slug such aswiento keep only one city. The Actor refuses any value that is not a slug, so check the spelling first. This run did not uselocation. -
Add full details when you need them.
fetchDetails: truefills the description, exact date, structured salary and employer website. Expect a slower run. This run did not use it. -
Filter for home office.
homeOffice: truekeeps only jobs the site marks as home office. This run used false, so it returned both kinds. -
Test with a small page count. Keep
maxPagesPerQueryat 1 while you develop. Raise it only when you need more jobs. -
Several keywords share one budget.
maxItemscaps the rows across all keywords together, so a long keyword list fills up the cap quickly. - Start with a broad search. A narrow keyword can return few or no rows. Check the count first, then tighten the search.
Troubleshooting common issues
The run ends FAILED with a message that names an input. Fix the input the message names, and run again. A bad location is the usual cause.
The run ends SUCCEEDED with zero rows and a status message. The Actor loaded the pages and found nothing that matched. The message suggests broader keywords or looser filters. This is a valid result, not an error, but your script must handle an empty dataset.
The run ends FAILED because the pages could not be read. The Actor could not load the site's pages, or their layout changed. The status message says so. Run it again later.
The script stops with a KeyError on APIFY_API_TOKEN. The variable is not set in the shell where you run the script. Export it there, then run the script again.
Next steps
-
Run the same script on a schedule with your own scheduler, and keep the run
idfrom each run to match results later. - Read the Apify Python client docs at https://docs.apify.com/api/client/python/ for the other calls the client offers.
-
Try a second keyword such as
javaorbuchhalter, and compare the rows.
For current pricing and details, see the Store page.
Top comments (0)