How Oink turns a job brief into explained matches

Oink is an AI job search assistant for people who want a short, reasoned list of roles instead of another infinite feed. It does not apply to jobs for you. It searches, compares each posting with what you have actually asked for, and gives you the original link so you can make the decision yourself.

1. You describe the job you want

The useful input is a real search brief, not a bag of keywords. Oink can work with your target role, relevant skills, seniority, preferred work arrangement, location or work eligibility, salary floor, and explicit dealbreakers. Telegram onboarding can start from those answers and optionally add a CV; the web onboarding currently uses a PDF resume as part of setup.

Preferences are treated as preferences. A past job on a resume is evidence that you can do something, but it is not automatically evidence that you want to do it again. If your brief says “staff backend roles, remote in Europe, no ad-tech,” those stated constraints should not be silently replaced by an older title in your work history.

For a practical way to write the brief, read how to describe the job you want to an AI. The goal is not perfect prose. It is enough specificity to distinguish a good role from a plausible-looking mismatch.

2. Oink searches more than one kind of source

Oink combines broad job boards, specialist and regional boards, employer career systems, monthly hiring threads, and open-web discovery. Some searches are shared across users and filtered later; others, including LinkedIn-derived and open-web searches, use search phrases derived from the current profile.

Source availability changes. A provider can be slow, blocked, empty for a query, or temporarily disabled by the operator. That is why Oink does not promise that every named source participates in every run. The job-sources guide lists the current categories and explains what can go missing.

Results are normalized into a comparable shape and deduplicated before matching. The original posting remains the authority: Oink links back to it, and you should confirm that the role is still open and that its terms have not changed before applying. Our guide to spotting stale and ghost listings explains the manual check.

3. Explicit conflicts are screened before nuanced matching

Some evidence is mechanical. A role that explicitly requires a location you cannot work from, a work arrangement you ruled out, or a clearly incompatible seniority should not consume the same attention as a close match with one uncertain detail. Oink uses source-grounded checks to identify explicit conflicts and then applies higher-quality contextual scoring to viable candidates.

The resulting match score runs from 0 to 5. Higher means the posting aligns more closely with the current brief. The score is a relevance ranking, not a probability of receiving an interview and not a guarantee that the employer will consider you eligible. Missing details stay missing; they should appear as uncertainty rather than be invented to make a role look stronger.

The AI job-matching guide explains the scale, what the reasons mean, and how to use a score without treating it as an oracle. If you have been applying at volume without replies, score-before-you-apply is the practical version of the same idea.

4. The strongest matches become actionable cards

Scheduled delivery ranks the eligible results and sends a bounded set of the strongest cards, subject to the user’s score floor. A Telegram card can include the title, company, location, original posting link, score, concrete reasons for fit, concerns or unknowns, and the structured job details that were actually available.

From the card you can open the posting, mark it applied, ask for a deeper fit analysis, or say “more like this” or “not for me.” Marking a role applied updates Oink’s tracker; it does not submit an application to the employer. Feedback can influence later scoring, but it should be understood as new preference evidence—not instant retraining after one tap.

The same matching pipeline supports the private web feed and a linked Telegram account. Telegram is the fastest way to receive the shortlist without keeping a browser tab open; the Telegram job-alert guide covers delivery behavior and its limits.

What Oink deliberately does not promise

• It does not guarantee that every source is healthy or returns results on every run.
• It does not guarantee that a third-party listing is current, legitimate, or still accepting applications.
• It does not turn a match score into an interview probability.
• It does not auto-apply or make an employment decision for you.
• It does not replace checking work authorization, compensation, or application terms at the original source.

That boundary is the point: Oink reduces the repetitive search and first-pass reading while keeping the consequential decision with the job seeker.