AI Hiring
How the AI interviewer works
You upload a resume, send Aara, our AI interviewer, into your Microsoft Teams meeting, and get a scored report for every answer. Here is exactly what happens at each step.
Upload the resume
Add the candidate's resume, and the job description if you have one. Questions are written from both.
Send Aara into Teams
Paste the Teams meeting link and launch, or schedule the interview for later.
Aara interviews
She asks the questions, listens, follows up on the answers and waits until the candidate has finished speaking.
Get the scored report
Every answer is graded, with the candidate's own words as evidence, and you decide.
1. Questions written from the resume and the job description
Upload the resume as a PDF, Word or text file. AI Hiring reads it and writes interview questions about the candidate's actual skills and projects. If you also upload the job description, the questions focus on what the role needs, and the skills the job description requires count double in the final score. You choose how many questions to ask and can set the role and experience level, so a fresher is not asked senior-level questions.
2. Aara joins the Teams meeting
Paste the Microsoft Teams meeting link and launch the interview. Aara joins the call like any other participant; you admit her from the lobby. You can also schedule an interview ahead of time: AI Hiring sends the candidate a reminder email and launches Aara at the scheduled time.
3. A real conversation, not a script
Aara asks one question at a time and speaks with a natural voice. She does not interrupt: she waits until the candidate has finished their answer before moving on. Her next question follows from the last answer: a strong answer gets a deeper follow-up, a weak one moves to a different skill, and after two questions on the same topic she moves on to another one.
Candidates can talk to her naturally. They can ask her to repeat or rephrase the question (she rephrases it without giving away the answer), ask for a moment to think, say they don't know, or ask to skip. You can follow the live transcript from your dashboard and pause, resume or restart the interview at any time.
4. Every answer graded, with evidence
When the interview ends, every answer is graded out of 10:
- Key points (6 marks): the points a strong answer covers, each marked covered, partly covered or missed, with a quote from what the candidate actually said.
- Depth (2 marks): whether they explained why or how, or gave a real example from their own work.
- Clarity (2 marks): whether the answer was easy to follow. Accent, filler words and transcription errors are never penalised.
A clearly wrong technical claim loses a mark. A correct answer that takes a different approach from the expected one still gets credit. A skipped question or "I don't know" scores 0.
5. The report, and your decision
The report gives an overall score out of 100, a score for each skill, a summary and a recommendation (Strong hire, Lean hire, Borderline or Not recommended). A candidate is never marked "Strong hire" while any skill is below 50%. The recommendation supports your decision; it does not replace it. A person on your team reviews the scores and makes the hiring call.
Coding assessments
For technical roles you can send a coding assessment after the interview: the candidate solves problems in the browser and you review the code and results.
How it compares
An interview copilot helps a person run the interview; a one-way video interview has no follow-ups. See the side-by-side comparison of the four ways to run a first round.