Keeps every candidate warm, scheduled, and told the truth.
Hiring is lost to latency more than to judgement: the good candidate who waited nine days for a reply, the panel feedback that never came, the rejection nobody sent. An AI recruiter fixes the parts that are pure diligence — screening against the scorecard you wrote down, scheduling, chasing, replying — and hands humans the judgement calls with everything already assembled.
The scheduling ping-pong, the feedback chasing, and the guilt of a candidate who never heard back.
- Application screenEvery weekday, 07:20
- Candidate repliesEvery weekday, 09:00
- Feedback chaseEvery weekday, 12:30
- Candidate stage healthEvery weekday, 17:30
- screen-against-scorecard
- schedule-interview-loop
- chase-panel-feedback
- write-candidate-reply
- warm-passive-candidate
What Noor does between waking the servers and going quiet.
Every hour below is one Routine you can open, read, and change — including the one where it stopped and asked a human.
- 07:20
Screens against the scorecard, not a vibe
Reads the eleven overnight applications against the criteria written down for this role, scores each against the same rubric, and records the reasoning so a decision can be argued with.
Bases - 09:00
Replies to everyone, including the no
Sends the invitations and the declines the same morning. A rejection at 09:00 on day one is a better candidate experience than an offer nine days late.
Email - 10:40
Ends the scheduling ping-pong
Finds the slot that works for a three-person panel and a candidate in another timezone, books it, and sends each interviewer the brief and the questions they own.
Workspace - 12:30
Chases the feedback nobody submits
Nudges the two interviewers who have not written up yesterday's loop, with the scorecard attached and the deadline visible. The candidate is not told 'we are still deciding' for a week.
Tasks - 14:45
Keeps the passive pipeline warm
Follows up with the four strong candidates from previous rounds who were not right then, with a reason to talk now rather than a template.
Email - 16:00
Leaves the judgement to a person
Stopped for youTwo finalists, different strengths, one role. It writes the comparison against the scorecard, states what it cannot judge, and puts the Decision in front of the hiring manager.
Decisions - 17:30
Keeps the pipeline honest
Updates every candidate's stage, flags the three who have been in the same stage for over a week, and reports where the funnel is actually losing people.
Bases - 19:00
Reports the week the way a person would
Applications in, screens done, loops run, offers out, and the one bottleneck that is slowing all of it down.
Workspace
- 0
- Candidates left without a reply
- 6
- Interviews scheduled, no back-and-forth
- 1
- Decision that needed you
- Two finalists, different strengths, one role. Here is the comparison — which offer do we make?
- This candidate is asking above the band. Stretch, or hold the range?
Answering one performs no side effect. The employee writes the question and the options itself, and an ordinary Member can answer it.
Capable in the ways that matter, bounded in the ways that count.
A pipeline that is a real table
Candidates, stages, scorecards, and attachments live in a Base with typed fields and saved views — worked by the AI Employee and by your hiring managers in the same place, not exported between two tools.
Screening you can audit
Every score cites the criterion it was given and the evidence it read. A rejection you disagree with can be opened, read, and reversed — which is the only version of automated screening worth having.
Scheduling across a panel
Finds the slot for four calendars and a timezone, books it, and sends each interviewer what they are responsible for. The candidate gets one email instead of six.
Nobody goes cold
A Routine watches for candidates stuck in a stage and for interviewers who owe feedback. Silence becomes a thing the system notices rather than a thing you feel bad about later.
Four documents, and then you stop being the trigger.
Everything that makes this a recruiter rather than any other role is plain, editable text. Change how it thinks by editing a document, the way you would rewrite a job description.
How it judges
- Voice, priorities, and the lines Noor will not cross alone.
What it repeats
- screen-against-scorecard
- schedule-interview-loop
- chase-panel-feedback
- write-candidate-reply
- warm-passive-candidate
When it works
- Application screen — Every weekday, 07:20
- Candidate replies — Every weekday, 09:00
- Feedback chase — Every weekday, 12:30
- Candidate stage health — Every weekday, 17:30
What it can reach
- Gmail Connection for the hiring mailbox
- Google Calendar Connection
- The candidate Base and its scorecards
- The role briefs and interview guides in Notes
It works your records, not a copy of them.
Everything Noor does happens inside the products your team already uses — the same rows, the same threads, the same queues. There is no export step and no second system of record.
See every productThe ones people actually ask.
- Is it making hiring decisions?
- No. It screens against criteria a human wrote down and shows its reasoning; the shortlist, the loop, and the offer are Decisions a person answers. Anything close to the line is escalated by design, not by luck.
- How does it avoid biased screening?
- It scores against your written scorecard and records the evidence for each criterion, so screens are reviewable and reversible rather than opaque. It is a diligence tool — the judgement stays with your hiring managers.
- Do I need an applicant tracking system too?
- For small teams, a Base with stages and scorecards is usually enough. If you already run an ATS, the AI Employee can work your mailbox and calendar alongside it.
Every other role, working the same way.
Build an autonomous company today.
Install Genosyn, choose an AI Model, write the first role, and put it on a schedule. Tomorrow morning, one job runs without you. The company grows from there.