New to this? You're in the right place. This post doesn't assume you know what "scraping" means, or how a computer could possibly "apply" to a job for you. If a word needs explaining, it's explained right where it shows up.
The problem this kind of software solves
Job hunting seriously means checking a dozen different job-board websites, over and over, for postings that match what you're looking for — then filling out largely the same application form, again and again, dozens of times. It's repetitive in a way that's genuinely exhausting, and it's easy to miss a good listing just because you didn't happen to check that particular site that week.
This kind of software automates the repetitive parts — checking the boards, filtering out anything that doesn't fit — and leaves the parts that actually need a person's judgment to a person.
The two words you'll see everywhere
- Scraping means having a program visit websites and copy information off them automatically — in this case, job postings. Picture a very fast, tireless assistant going through every listing on a job board and writing down only the ones worth a second look, instead of you doing that by hand.
- Auto-apply means the program fills out a job application form for you — your name, your resume, the standard questions. It does not mean the program decides on its own to apply and hits send.
What gets automated — and what doesn't
Here's the part worth being completely direct about, because "auto-apply" can sound like applications go out without anyone checking: a person reviews and approves every single job before anything gets filled out, and a person makes the final click on every submission. The pipeline looks like this:
▶ show code▼ hide code
Scrape → Match → Review → Auto-Apply → Done
The program handles the first and last technical steps — finding listings and filling out forms. The middle step, deciding which jobs are actually worth applying to, stays a human decision the whole way through. And even after you approve a job, the program fills out the form and then stops and waits — it never submits on its own. You always make that final click.
What you'd actually see
You'd review a simple list — a spreadsheet, or a similar tracked list if you're using it inside a project-management tool — with each job marked Yes, No, or Maybe. That's the entire manual workload most of the time. Everything before that decision (finding and filtering listings) and everything after it (filling out the form) happens without you having to do it by hand.
Behind that simple list, quite a lot is actually being checked: does the salary meet your floor, does the location and work type fit, does the listing avoid your exclude-list keywords (so a "Senior Director" role doesn't show up if you're not looking for one). All of that filtering happens the same way every time — no guessing, no AI making a judgment call about whether a job "feels right."
What this isn't (yet)
An honest note, because overselling doesn't help anyone: this can't get past every application form on its own. If a form asks an unusual custom question, or shows a CAPTCHA, it stops and flags that application as needing you to finish it by hand — it doesn't guess, and it doesn't get stuck pretending to make progress.
Ready for the deeper version?
Everything above describes the shape of the Job Hunter open-source project, which also has a matching template you can import into OpenTechnologyApp if you'd rather track things in queues and dashboards than a spreadsheet. If you want the real technical detail — every job source, exactly how the matching works, how to run it yourself — the full version is here: Job Hunter: Open-Source Job Search Automation That Actually Applies.
If this was your first time reading about job-search automation, you now know enough to follow that post — and enough to tell whether this kind of tool would actually help your search.