Screening basics
What AI pre-screening is, and when to use it.
A first interview that candidates take on their own time. Software asks the questions and writes up the answers, so your team starts with evidence. Here’s how it works.
6 min read · Updated September 24, 2026 · The LessRounds team
AI pre-screening is a first interview that a candidate takes on their own time. Software asks the questions and writes up the answers, so your team starts with what each candidate actually said instead of a pile of résumés. It happens after someone applies and before anyone on your team talks to them. This guide explains what it is, how it compares with the two things it usually replaces, and when it’s the wrong tool.
What AI pre-screening is
The candidate gets a link. They open it on a phone or laptop whenever suits them, answer a few questions out loud, and leave. Software asks the questions, listens, and writes up what it heard against the skills you said matter for the role.
Most AI pre-screening interviews have three parts:
- Screening questions. Short factual ones: which shifts can you work, when could you start, are you based in the city or willing to move.
- A spoken interview. Three to eight open questions about the job, answered on video or audio. Some tools read the questions word for word. Others hold a conversation and ask follow-up questions when an answer is short or vague.
- A write-up. A score per skill, the reasoning behind it, and the lines from the transcript that support it.
The important word is “pre”, meaning “before”. It doesn’t decide who gets the job. It helps you choose which candidates someone on your team should talk to next, and it gives that person something concrete to read first.
How it differs from résumé screening
A résumé tells you what someone says they’ve done. An AI pre-screening interview shows you how they think and explain things when asked.
That difference matters most for candidates whose résumés undersell them: students, career changers, people applying for hourly roles, and anyone who has never had help writing one. Résumé screening quietly rewards people who are good at résumés. A spoken first interview gives everyone the same questions and the same chance to answer them.
Résumés are still useful. They’re a fast way to check hard requirements like a license or a degree the role truly needs. Many teams do a quick résumé check and send the AI interview to everyone who passes it, rather than trying to pick the best ten from paper alone.
How it differs from a phone interview
A phone screen is a short first interview by phone. It’s a live call between a recruiter and a candidate, usually 15 to 30 minutes long. It’s good at two things software can’t do: answering the candidate’s questions and making the job sound attractive.
It’s weak when you need every candidate treated the same way, or when you have a lot of candidates. Each call depends on who made it, what they asked, and how tired they were by the eighth call of the afternoon. It also needs a time when both people are free, and finding one can take days.
| Reading résumés | Phone interview | AI pre-screening | |
|---|---|---|---|
| What you learn | The experience they say they have | Their answers and how they come across | Answers to set questions, with evidence |
| Your team’s time per candidate | A minute or two | 20–30 minutes plus scheduling | A few minutes to read the write-up |
| When it happens | Whenever you get to it | When both calendars allow | Whenever the candidate chooses |
| Same for everyone | Depends on the reader | Depends on the caller | Same questions, same scoring |
| What it misses | How people think | Candidates you never had time to call | A back-and-forth conversation |
In short: with AI pre-screening, you give up the back-and-forth of a live call. In return, you hear from far more candidates and ask all of them the same questions. You get the conversation back in the next step, with a short call to the people you already know are worth it.
When it works well
- You get more applicants than you can call. Jobs with lots of applicants, hiring straight from colleges, and internships are the obvious cases.
- The skills can be heard in a conversation. Explaining, handling an upset customer, reasoning through a scenario, describing how they’d prioritize a busy day.
- Candidates can’t take calls during office hours. People working shifts, students in class, and applicants in other time zones.
- You want to hear from every applicant, including the ones a résumé skim would have skipped.
- You hire in more than one language, and want candidates to answer in the one they’re most comfortable in.
When not to use it
- A few senior candidates you’re trying to win over. A call from the hiring manager will do more than a link.
- Skills that have to be seen. Driving, lifting, operating equipment or cooking need a practical check. An AI interview can cover talking about the job, not doing the job itself.
- You can’t review results within a day or two. Interviews that candidates take on their own time don’t remove the waiting. They move it from the candidate to your inbox. If nobody reads the write-ups, you’ve just added a step.
- You plan to let the score decide. A score helps you decide who to look at first. It isn’t a final decision. Someone should read the evidence behind it.
- People you’ve already met, such as strong referrals or returning employees. Send them straight to the next interview.
A simple rule of thumb: if the first conversation with a candidate is mostly the same questions every time, and you have more candidates than time, AI pre-screening will help. If every first conversation is different because every candidate is different, keep that first conversation with a person.
What to check before you pick a tool
Ask these questions of any product, including ours:
- Does every score show its evidence, or only a number?
- Can you choose the skills and the level you expect, or does it score a generic “fit”?
- What does it judge: the words candidates say, or their faces, voices and backgrounds?
- Does it email candidates rejections on its own?
- Can candidates take it on a phone, in the language they’re most comfortable with?
- What happens to candidate data, and is it used to train AI models?
- How is it priced: per user, per month, or per interview?
LessRounds was built around those answers. It scores only the written transcript of what candidates say, shows reasoning and a quote for every skill, never emails candidates a result on its own, and runs interviews in English or Hindi. How it works walks through the whole flow in a few minutes.