Straight answers.

No. An ATS tracks applicants through a process. notanATS is the screening layer: it holds the hiring manager's definition of the work, screens applicants against it, and moves the right people to the interview. It can run in front of an ATS you already use — the Workable integration is live, and Greenhouse, Lever, Ashby and SmartRecruiters are coming soon — or on its own, if you do not have one.

No. notanATS screens; it never interviews. The screening is a typed chat about the work. The interview is the real, booked conversation with your team, and it stays that way by design.

Calibration is how a position's screening model gets defined. notanATS drafts the model from the job description first — dimensions, what good looks like, what capability looks like — then a guided chat walks the manager through it: confirm, correct, cut, mark the must-haves, lock. It is built to be finished in one sitting, and most of that sitting is confirming a draft rather than writing from a blank page. Only the hiring manager can calibrate. Not an assistant, not an agent.

Yes. After real interviews, the manager's interview notes can amend the model into a new version — a new anti-pattern, a stronger signal, a re-weighted dimension. Amendments apply to future applicants only. Every applicant keeps the version they were screened against, pinned to their record.

On evidence. When a screening chat ends, the applicant's answers are matched against each dimension of the locked model and rated. The ratings combine by the weights set at calibration into a score out of 100, and the per-dimension detail is kept alongside the total. If a must-have dimension is missed, the application is flagged — visible to the manager, skipped by automatic advancement, but never deleted or hidden.

They cannot see the rubric, because the screening never has it. Applicants do see what the work is and what capability looks like, up front — knowing what the job involves is not gaming; it is the point.

A screening about the actual work, in their own words, with no account to create and no tracking. And an ending: when the position closes, applicants hear back.

Create a position, hand you the calibration link, read the pool, and advance the applicants you choose. Two things it can never do: calibrate — defining the work is yours alone — and screen, because notanATS is the screener. The agent is a control channel. The full picture is on the agents page.

Any agent or tool that can call an API with a key. Access runs on an organisation API key — shown once at creation, stored as a hash, scoped to your organisation, revocable at any time.

Selection rates are monitored with the EEOC four-fifths rule and flagged when a group's rate falls below 80% of the highest group's rate. The trust page describes the mechanisms in full.

Names and emails are masked out of what is screened and scored. Applicant pages carry no analytics. Screening conversations are purged after 30 days on free plans; on paid plans, applications are anonymised after 12 months. Applicants can request erasure. Details are on the privacy page.

The application goes through anyway. A resume that fails to parse never blocks an applicant — the screening is a conversation about the work, not a reading of the document.

Starting is free; after that you pay per completed screening, invoiced monthly after the fact. The pricing page explains the model. Exact rates for your volume are confirmed when your account is set up.

You, if you do your own screening — a founder hiring a first team, an operations lead with no recruiter, a manager whose ATS tracks everything except whether someone can do the work. If screening is the part of hiring that eats your week, this is the part notanATS takes over.