CiiVSOFT reads CVs so your recruiters don’t have to, but it never keeps them or looks at PII. No candidate data stored. No AI model trained on your information or exposed to PII. Every decision still made by a human on your team.
Job details and candidate CVs, retrieved only through the secure connection your team controls, and revocable by you at any time.
Almost nothing. A Job ID, a Candidate ID, an Application ID, a timestamp, and the evaluation output, written straight back into your ATS. No CVs. No names. No contact details. No original documents, ever.
CiiVSOFT gives your recruiters a faster first pass. It doesn’t replace their judgment, and it doesn’t quietly build a data footprint behind the scenes. Everything it produces is written into your own ATS, in plain view, for your team to review and decide on.
We read job and candidate data directly from your ATS, through access you grant and can revoke at any time.
Personal details are stripped out immediately, before anything reaches our AI.
The redacted content is compared against the job’s screening criteria. No candidate is ever compared to another.
The result goes straight into your ATS. CiiVSOFT keeps only a technical record of what happened: never the CV, never the candidate’s details.
Data is encrypted at rest and in transit, with keys rotated regularly.
Active sovereign data regions (EU, US), with additional regions available on request.
CiiVSOFT holds UK Government-backed Cyber Essentials certification.
Our infrastructure has no public-facing login or portal; access runs entirely through your own ATS.
Our AI is built to disregard protected characteristics under the Equality Act, and every output remains open to review.
Powered by foundation AI models from Anthropic and Amazon, accessed through AWS, never used to train on your data.
CiiVSOFT connects only through your ATS, using access your team grants, and can revoke at any time.
Every evaluation is written back to your ATS for your team to review; the hiring decision is always a person’s.
We share full documentation, including data processing terms, security posture, and sub-processor list, as part of onboarding, and we’re happy to walk your security team through it directly.
We only use these details to send your security pack and follow up, in line with our Privacy Policy.