Loading…
Kirnova Ask
Ask a plain question about your people and get an answer from your own records on Kirnova.
Who it's for
Every hirer role in a company, recruiters over the candidates on their desk, and college staff over their own students. Candidates and Kirnova's own admins can't use it. You only ever ask about people you already have a route to.
It reads readings, applications, stages and teams on Kirnova. It doesn't connect to your HR system, your mailbox or your spreadsheets.
What you can ask
Type in your own words, or press one of the questions the box suggests for the page you're on. On a job, it answers about that job's applicants. On a person, about them. On a team, about the team.
About a job
Who hasn't started TIGER Core, which reports are ready, who is shortlisted, how many applied, and everyone's readings.
About a person
A short summary, interview questions, their first 90 days, how to manage them, and what to probe before an offer.
About a team
How the team reads on one reading, where nobody sits, and two people side by side.
How it works
Most questions are read with fixed word tables, like ordinary software. A small outside model reads only the hard ones, and writes a one-line summary that must pass the same wording check as a report. Names are replaced with placeholders before any model sees a word.
People are put in order only on one named reading. Where readings overlap, they sit level. Thin evidence is placed last as "Not enough evidence to place". Every sentence about a person comes from the same content bank that writes the TIGER report, and every answer says what it was based on. It never makes a reading or a decision, and it never presses a button for you.
Ask Kirnova · Sales Executive, Pune
Asking about the 7 applicants to Sales Executive, Pune.
- Who hasn't started TIGER Core
- Which reports are ready
- Who is shortlisted
- Execution across my candidates
Who on this job is strongest on Execution?
Two of your four finished reports read high on Execution, and they sit too close together to put one above the other.
How your candidates read on Execution.
High on ExecutionSolid reading
- Meera Kulkarni, finishes what she starts, says so when a date will slip
- Aarav Shinde, steady on long work, slower to ask for help
Moderate on ExecutionSolid reading
- Rohan Deshmukh, strong start, needs a check-in near the end
Not enough evidence to place
- Kabir Nair, the answers did not hold together, a free second attempt is offered
Why there is no single order: Meera and Aarav read within the same range on Execution, so they are shown as level. Kirnova draws no order on thin evidence.
- 3 applicants have no TIGER Core reading that you can see.
Based on the 4 TIGER Core reports you hold for this job. Nobody's TIGER Light was read.
Limits
Daily limit
100 questions a day per person. After that, nothing is sent to the model.
Large lists
It shows the 500 most recent people. Narrow the question to a job for the rest.
Your role
The box has the same permissions as the screens. A question your role can't ask doesn't appear.
Privacy
A conversation is visible only to the person who asked it. When a person is erased, every question about them goes too.
Outside companies
Each one is named on our subprocessors page.