How it works
Turn candidate claims into evidence.
Three steps, typically under a minute. You stay in control of every judgement along the way.
Add the candidate
Paste their LinkedIn profile and GitHub profile and upload their CV.
- CVs are accepted as PDF or DOCX and stored privately against your account.
- GitHub is optional — plenty of excellent engineers have no public activity at all.
- Profile information is only ever used when you supply it. Verichi does not scrape LinkedIn.
Verichi cross-references
Verichi compares employment history, technical experience, projects, dates and professional claims.
- Employment rows are matched company by company, title by title, date by date.
- Technical skills are checked against every supplied source, including public repositories.
- Findings are categorised as supported, differing, or simply not evidenced.
Know what to investigate
Receive a consistency score, supporting evidence, potential discrepancies and targeted interview questions.
- A transparent, rules-based score — no black-box model deciding who is credible.
- Every flag explains why it was raised and what evidence exists.
- Interview questions are written to open a conversation, not to corner anyone.
How the score works
The Candidate Consistency Score is rules-based and deliberately simple. Every candidate starts at 100. Points are deducted only for evidence-backed inconsistencies between the sources you supplied.
−5
Minor discrepancy
−10
Moderate discrepancy
−15
Significant discrepancy
−20
Major contradiction
No public evidence never reduces the score. If a CV lists eight years of Python and there are no public Python repositories, Verichi records “public GitHub evidence not available” — not a concern.