Before you book the interview, make the evidence earn it.
Lever’s AI Screening by VONQ adds structured chat or voice conversations with explainable results. Choose Cernor when you need to establish what each career supports before asking more questions. It tests important gaps against your agreed role requirements, called the Standard, and shows what the answers mean for the next hiring step.
No credit card required.
Start instantly with a job description and resumes. From $199 per live role, with unlimited team members and no monthly subscription.
Cernor vs Lever AI Screening Updated September 2026 First-party sources
One role. Three candidates. Questions that fit the evidence.
All three are read against the identical locked Standard. What Cernor asks each of them is derived from what their own record failed to settle, which is why the three columns below do not match, and why that is the point rather than an inconsistency.
This example shows the Cernor workflow: supported evidence stays in view, material gaps receive targeted attention and Trust questions stay distinct from capability. Every answer is evaluated against the same role criteria.
Polish gets attention. Evidence should get the interview.
A fluent account of a project can hide an important ambiguity: what did this person actually own? Cernor reads the career in context and tests consequential claims through questions about scope, decisions and constraints.
Fit, Evidence and Trust remain distinct. A strong career can still contain a claim worth checking. Follow-Up gathers more evidence, and the recommendation moves only as that evidence warrants.
Ten questions, and where the two answers stop agreeing.
| Question | Cernor | Lever |
|---|---|---|
| Does every applicant get read? | Yes: the whole cohort against one agreed set of role requirements | Chat and voice screening evaluated against the role criteria |
| Where do the criteria come from? | The hiring team authors and locks them before the first candidate is read | The screening experience is composed for the role |
| Who gets asked questions? | Only candidates whose own record left something decision-critical unresolved | Candidates entering the AI screening workflow |
| Where do the questions come from? | That candidate’s specific unresolved claim, with each tier depending on the answer before it | Job-relevant conversations based on the role requirements |
| Do two candidates on the same role get the same questions? | The path follows each candidate’s material evidence gaps | The product page does not specify identical questions for every candidate |
| What happens to a polished claim the career does not support? | Fit, Evidence and Trust are judged separately, so an impressive record carrying an unsupported claim is its own finding | An explainable fit score with data-backed justification |
| Can the answer move someone up the shortlist? | Yes: the returned path is judged against the same Standard, and the movement is attributable to the answer that caused it | Screening responses inform the fit score and prioritisation |
| What happens if the hiring manager changes the requirement mid-search? | The Standard versions; only affected candidates reopen, and you see who moved and who held | Not described in the current public material |
| Does it work with our ATS? | Yes: the ATS stays the system of record; connecting one is optional to start | Lever is the ATS and CRM |
| What does the hiring manager receive? | A Decision Packet: who to interview, what the evidence established, what is still open, and what moved | An explainable score highlighting candidate fit, a structured summary and the full interview transcript |
A consistent screening layer, built to be defended.
Lever’s AI Screening by VONQ uses structured chat or voice conversations, evaluates responses against role criteria and returns explainable scores, summaries and transcripts. It puts additional screening information into the hiring workflow.
Every question should serve the hiring decision.
Consistency belongs in the bar
Every applicant is evaluated against the same approved Standard. Your team can inspect the requirement behind a finding and preserve the basis for a decision when the role changes.
Build on the evidence you already have
The existing record is read first, so the system knows per person which criteria are already settled. A candidate who evidenced everything material is asked nothing at all, and that silence is a result, not an omission.
The question comes from their record
For everyone else the path starts at their own unresolved claim, the next question depends on the last answer, and a later tier can push on scope, consistency or proof.
Different questions, still comparable
Different questions can produce evidence against the same role criteria. Cernor records what each answer establishes and what remains open, so your team compares candidates on the work the role requires.
Hand over something usable
The hiring manager gets who to interview, what the evidence established, what is still open and what changed: plus, for each person, why they were asked what they were asked.
Keep Lever. Connect Cernor.
Lever
Pipeline, CRM, nurture, reporting: the system of record, unchanged.
Cernor
The deep read on the candidates already in it: what is settled, what is unresolved, who is worth asking and what the answer changes.
The join
Cernor supports ATS-connected workflows, and the Trial does not require one: bring a job description and the resumes you already have.
What buyers actually ask.
How does Cernor keep different questions comparable?
The role criteria stay fixed. Cernor asks about the evidence missing from each record, then evaluates the answers against that same Standard. Fit, Evidence and Trust remain visible, so a different question does not mean a different bar.
Is Cernor replacing Lever?
Lever stays your ATS and CRM. Cernor supports ATS-connected workflows and gives the team a deeper assessment of the candidates already in the pipeline. You can start without connecting an ATS.
Why add Cernor if we already use AI Screening?
Choose Cernor for the complete decision record: the criterion, supporting evidence, open question, returned answer and resulting recommendation. That gives your team a clear account of who deserves interview time and what still needs checking.
What can our team inspect after the decision?
The criteria in force, evidence behind the findings, unresolved claims, Follow-Up and what its answers changed. Cernor preserves that context alongside the recommendation, including when a changed requirement reopens affected findings.
Published by Cernor, and checkable line by line.
Lever AI Screening by VONQ checked against first-party product material on 7 September 2026. Capabilities described here are published claims, not results from a comparative product test.
Give the interview to the evidence. Start with Cernor.
One live role, read against a bar you set first.
No credit card required.