Rank the whole pile against one rubric, with the line from the CV behind every score. Then match, write, book and measure — without leaving.
The HireOS workspace, rebuilt from the application’s own components. Illustrative data. How the scoring works
The recruiter’s day
The platform
Not seven tools you integrate. One workspace you switch inside — shortlist someone in Screen and they are already waiting in Outreach and Interview.
Buy the ones you want. Every module is a peer and none of them is a mandatory base.
Rank a pile of CVs against a job description, with the evidence behind every score.
Available nowScore a new role against the people you have already screened, before you advertise it.
Available nowDrafts the approach, the chase and the rejection. A recruiter approves before it sends.
Available nowSlots, invitations and reminders, without the six-message thread to find an hour.
Available nowWhat each client actually wants, learned from the roles you have filled for them.
Available nowTime to shortlist, submission to interview, and which roles are quietly stalling.
Available nowOne instruction. It writes the role, screens the pile and books the interviews you run.
Available nowSee them on your own role
Twenty minutes, your job description, your CVs. A 7-day trial after.
Book a callInside each one
AI screening
Available nowPaste the job description, drop in the pile, and read a ranked shortlist with the reasoning already written.
One rubric, built before any CV is read
The job description becomes a fixed list of weighted requirements, so every candidate is measured against the same yardstick instead of criteria that drift as the batch goes on.
Evidence, not vibes
The model judges each requirement met, partial or missing — and every “met” has to carry a verbatim quote from the CV. Confident wording earns nothing.
Scores computed in code
The number is arithmetic over those verdicts, not a figure the model invented. The same CV always scores the same, and a missing must-have caps the result instead of being averaged away.
How it fits together
Paste the JD, drop in the CVs.
Score the pool you already have.
Drafts you approve, then send.
Offer times, confirm, send it.
Why the ranking holds up
The job description becomes a weighted list before any CV is opened, so criteria cannot drift by candidate seventy-four.
Every “met” carries a verbatim quote from the CV. Confident wording earns nothing on its own.
The score is computed in code from those verdicts, so the same CV always scores the same.
Your specialism
Technology & Engineering
Two hundred applicants for one backend role, most of them keyword-matched and none of them read.
Where the line is
Throughput is the part that eats the day. Judgement is the part your clients pay you for — and HireOS does not touch it.
Every requirement carries a verdict and the line from the CV it came from. That is what makes a shortlist defensible to a client, and what lets a recruiter check a score in five seconds instead of trusting it.
How the scoring worksThe whole pile stays ranked and visible. The failure mode of automated screening is not a bad ranking — it is a good candidate deleted before a person saw them.
Where the line isOutreach drafts from the actual CV, not a template with a merge field. A recruiter reads, edits and sends it from their own mail client.
How Outreach worksThe person who scored 71 on a role they did not get is often an 89 on the one that just came in. Match scores them before you spend anything advertising.
How Match worksDeletions, exports and retention purges are recorded in an append-only trail that cannot be edited from inside the product.
How we handle dataPricing
Everything a desk needs
Any module, on its own
$75/ month
All seven modules
$379/ month
Book a callA 7-day trial on one of your live roles comes first. No card, and no subscription until it has earned one.
Questions
The offer
Take a vacancy you are working right now and the CVs already sitting in your inbox. Compare the shortlist to the one your recruiter would have built.
