If you’re job hunting, or keeping an eye on who’s hiring for work reasons, you’ll know the routine: open a dozen careers pages, scroll, forget which roles you’ve already seen, repeat next week. Most companies don’t host those pages themselves. They use an applicant tracking system (ATS) such as Greenhouse, Lever, Workday, Ashby, SmartRecruiters, Workable, Recruitee, Personio or Teamtailor, and many of these publish job listings through public feeds so that jobs can appear on careers sites and job boards.
Greenhouse, for example, documents its Job Board API openly and states that job board data is publicly available, so no authentication is needed to read it. That makes it possible to gather jobs from many employers into one tidy spreadsheet, and then hand the boring sorting to an AI assistant.
Quick answer: Collect open roles from the employers you’re interested in into a single CSV file, with the same columns for every job. Upload the file to an AI assistant that accepts files, such as ChatGPT, Claude, Gemini or Copilot, and ask it to shortlist roles that match your skills, location and level, summarise who’s hiring and turn long adverts into short summaries. Ask it to show its counts, and always open the original advert before applying.
New to AI? Start with our complete beginner’s guide.
Step 1: Collect the jobs in one file
We make the tool used here: ATS Jobs Scraper on Apify. You paste a list of careers page links (or company names), it works out which of the nine platforms each company uses, and it returns one row per job with the same columns every time: company, platform, title, department, location, a remote flag, workplace type (remote, hybrid or onsite), employment type, posting date, salary where the employer publishes it, the job link and the description. You can then download the results as CSV or Excel.
Useful options:
- Keywords and exclusions, e.g. keep “data” and “analyst”, drop “intern”.
- Locations, e.g. “London” or “remote”.
- Posted within N days, to ignore old adverts.
- Only new jobs, which remembers what you’ve already seen so a weekly scheduled run shows only fresh roles.
It runs on Apify and is paid per result (at the time of writing, $2 per 1,000 jobs plus $0.002 per run), so check the current price on the page before you start.
What a real run looks like
On 21 October 2026 we ran the tool’s default example: four employers on four different platforms, capped at 25 jobs per company.
| Company | Platform | Open jobs found | Jobs kept (cap 25) |
|---|---|---|---|
| Airbnb | Greenhouse | 192 | 25 |
| Spotify | Lever | 79 | 25 |
| Linear | Ashby | 29 | 25 |
| Hugging Face | Workable | 6 | 6 |
That’s 306 open jobs found and 81 kept, in about two seconds. Of those 81 rows, 35 were marked remote, 13 hybrid and 8 onsite (the other 25 had no workplace type in the source data), 13 mentioned London, and 29 had been posted in the previous 30 days.
Real-life example: The same day, a separate test of the “only new jobs” mode covered 11 employers across all nine platforms (including Stripe and Figma on Greenhouse, Zendesk on Workday, Canva on SmartRecruiters, bunq on Recruitee, 1KOMMA5° on Personio and Polestar on Teamtailor) and recorded 1,749 job IDs in one go. Numbers like these change daily, so treat them as a snapshot.
Free download: sample CSV from the 21 October 2026 run (81 real job rows; descriptions shortened to a 400-character excerpt). It’s handy for trying the prompts below before you run anything yourself.
Step 2: Let AI filter and summarise
Most AI assistants that accept file uploads (ChatGPT, Claude, Gemini, Copilot) can read a CSV like this. Upload the file, then try these prompts.
1. Shortlist roles that fit you
I've uploaded a CSV of job openings. Columns include company, title, department, location, remote, workplace_type, posted_at and description_excerpt. I'm looking for [ROLE, e.g. data or analytics] roles, based in or remote from [LOCATION, e.g. the UK], at [LEVEL, e.g. mid to senior] level. Filter the file and give me a table of the best 10 matches with company, title, location, workplace_type, posted_at and job_url. Add one line on why each one fits. Only use information that is in the file; if something isn't stated, say "not stated".
2. Get a quick overview of who’s hiring
Using the uploaded jobs CSV, summarise hiring by company: number of roles, the top three departments, how many are remote, hybrid or onsite, and the newest posting date. Then write three short bullet points on patterns you notice. Show your counts so I can check them.
3. Turn a long advert into something readable
Here is a job advert: [PASTE THE FULL DESCRIPTION FROM THE JOB LINK]. Summarise it in five bullet points: what the role does, must-have skills, nice-to-have skills, location and working pattern, and anything about pay or benefits. Then list three questions I should ask in a first interview. Don't add anything that isn't in the advert.
Found a role worth applying for? Use AI to tailor your CV, write a cover letter and prepare for the interview.
Tips for good results
- Check the counts. AI tools can miscount rows, so ask them to show their working and spot-check in a spreadsheet filter.
- Open the job link before applying. Adverts close and change; the link takes you to the employer’s own page.
- Keep it focused. A tight company list and a keyword filter give a shortlist you’ll actually read.
- Respect the data. These are public adverts, but use them for your own search or research, not for spamming recruiters.
FAQ
Do I need the tool to use the prompts?
No. The prompts work on any spreadsheet of jobs, including one you build by hand by copying titles, locations and links from careers pages. The free sample CSV lets you try them straight away.
Can AI tell me which jobs I’ll get?
No. It can only compare what’s written in the advert with what you tell it about yourself. Use it to save time sorting, and make the decisions yourself.
Try it yourself with ChatGPT
Here is a job advert: [PASTE THE FULL DESCRIPTION FROM THE JOB LINK]. Summarise it in five bullet points: what the role does, must-have skills, nice-to-have skills, location and working pattern, and anything about pay or benefits. Then list three questions I should ask in a first interview. Don't add anything that isn't in the advert.
Get the free AI Agents Playbook
50 practical AI workflows, plus one practical how-to a week in your inbox.
Free. No spam. Unsubscribe any time.
Related guides
- AI for Job Hunting in the UK: CV, LinkedIn, Interviews and Offers
- How to Use AI to Improve Your LinkedIn Profile
- How to Use AI to Plan a Career Change
Reference: Greenhouse: Job Board API documentation