Faculty rank
- Regents Professor of ChemistryDistinguished Professortenure-track
- Smith Family Endowed ChairProfessortenure-track
- Research Assistant Professornon-tenure-track
Verified academic records, inside your platform.
CVParsa IntegrationsLicensed to platforms and institutions
The engine behind every Cevian Labs product, connected to your own system through a secure API. Each CV comes back as one structured record, classified on academic vocabularies and checked against the public record.
Nine data classesin every record, or only the ones you need
9,000+ honorsin the Cevian Labs Academic Honors Registry
One record per CVjoined to your data by persistent identifiers
For systems that workwith academic careers.
Hiring and talent platforms
Give recruiters and search committees a comparable record for every academic applicant, instead of a stack of CVs in different formats.
Research-information systems
Populate faculty profiles with publications, grants, and honors drawn from the CVs faculty already keep, each entry checked against its source.
Universities and institutions
Connect the engine to your faculty-reporting system: faculty upload their current CV, and the engine identifies their recent activities automatically, with consistent records across departments and campuses.
Tailored toyour platform.
The engine reads an academic CV in any layout and discipline, from any country, and returns one structured record of that person's scholarly career. We tailor which CV data it extracts, and which of that data it checks against the public record, to what your platform needs.
- Choose the dataTake the full career record, or only the data classes your product uses.
- Choose what is verifiedPublications, grants, and honors can each be checked against the public record, or returned as the CV states them.
- Collect only what you useData your platform doesn't need is never extracted.
An example configuration
What one partner might choose to extract and verify
| Data class | Extract | Verify |
|---|
It understands academia.
The engine doesn't simply lift strings off the page. It classifies every entry onto closed academic vocabularies, so your platform can filter, count, and compare records across institutions, fields, and countries without hand-coding.
A few examples of what it standardizes and classifies.
Degrees
- PhDPHDDoctor of PhilosophyPh.D.research doctorate
- BA (Hons)B.A.undergraduate, with honors
- MBAM.B.A.professional master's
Advising
- Dissertation co-chairPh.D. advisorDoctoral advisor
- Honors thesis advisorUndergraduate advisor
- Ph.D. and M.S. advisorTwo recordsdoctoralmaster's
Career roles
- Chair, Department of HistoryAdministrative
- Postdoctoral Research AssociatePostdoc
- Senior Research ScientistResearch staff
Publications
- Proceedings of the Annual Conference on …Conference paper
- arXiv preprint, 2023Preprint
- Journal of …, 12(3):45–67Journal article
Funding or honor?
- Outstanding Teaching AwardHonornot funding
- NIH K99/R00 Pathway to Independence AwardGrantcareer-development fellowship
- NSF research grant, $450,000Grantresearch grant
Grants
- Principal Investigator, NSF, $450,000Research grantPI
- Sloan Research FellowshipFellowshipindividual
- NIH T32 training program (Director)Training grantinstitutional
Checked against the public record,with provenance on every entry.
Publications are matched against Crossref and other authoritative bibliographic and library databases; grants against NSF, NIH, USASpending, UKRI, and the EU's CORDIS; and honors against the Cevian Labs Academic Honors Registry of more than 9,000 honors from more than 4,000 organizations.
Each entry names its source
Peter Liebman, from our sample pool, pseudonymised
- PublicationCollective dynamics of 'small-world' networks, Nature, 1998
- GrantNSF award 3140822, $1,250,000
- HonorSIAM Fellow, 2009
Because every entry says which authority confirmed it, or that none did, your platform can show what an independent source corroborates, compute that share for each CV, and filter, weight, or audit on it. Checks are tuned for precision: an entry no source confirms keeps the CV's own content, is labeled unconfirmed, and is never matched to a guess.
Built to joinyour data.
Classification fields draw on closed vocabularies, so they filter and aggregate directly. Institutions, publications, grants, and people carry persistent public identifiers (ROR IDs, DOIs, funders' award numbers, and ORCID iDs), so records join cleanly to the institutional, bibliometric, funding, and researcher-profile systems you already run.
A few of the fields you can filter on directly
- Rank: Distinguished Professor
- Tenure: tenure-track
- Degree: research doctorate
- Role: postdoc
- Publication: journal article
- Funding: fellowship
- Honor scope: international
One record, joined four ways
Peter Liebman, from our sample pool, pseudonymised
InstitutionNorthwestern UniversityROR 000e0be47
Institutional data
PublicationCollective dynamics of 'small-world' networksDOI 10.5555/scn.1998.0193
Bibliometric databases
GrantFoundations of Data-Driven Dynamics (co‑PI)NSF award 3140822
Funder records
PersonPeter J. LiebmanORCID iD 0000-0004-5174-2936
Researcher profiles
How an integration works.
Talk to us
Tell us about your platform, your users, and the academic data they need.
Scope it together
We agree which CV data the engine extracts and verifies for you, and how records reach your system.
License and connect
Your platform connects to the engine through a secure API, under a license that fits your product.
See it work on a real CV
One CV's journeyFrom its first page to the finished, verified record. Plays automatically.
Click the slides or use the arrow keys to step through at your own pace. Open full screen · Full walkthrough
Why not just ask ChatGPT or Claude?Reading a CV is the part general-purpose chatbots already do well.
| Task | General-purpose chatbot (ChatGPT, Claude) | CVParsa |
|---|---|---|
| Read one CV and summarize it | Does this well | Does this well |
| Apply the same fields and vocabulary to every CV in a pool of hundreds | Output varies between runs | One schema, fixed academic vocabularies |
| Know that a fellowship title can be a postdoc, and that a CAREER award is both an honor and a grant | Inconsistently | Applied the same way to every CV |
| Count publications | Depends on how the question is asked | Publications counted by type; preprints and patents kept separate |
| Confirm each record against Crossref, NSF, NIH, and Cevian Labs databases | Not systematically, and prone to inventing citations it cannot find | Every record states which authority confirmed it, or that none did |
| Handle confidential applicant files | Consumer accounts raise data-governance questions | Encrypted, never used to train AI models, originals deleted within 30 days |
Confidential by design.
EncryptedAES-256 at rest, TLS in transit.
A secure APIAccess to the engine is authenticated, encrypted, and limited to licensed partners.
Never used for trainingSubmitted files are never used to train third-party AI models.
Handling set in your licenseRetention and data handling are agreed in each license. By default, original files are kept at most 30 days.
Questions.
Is there an API?
Yes. Licensed partners connect to the CVParsa engine through a secure API. Access and delivery are agreed as part of each license; there is no public, self-serve API.
For your engineers. A REST API over HTTPS. Your system authenticates with credentials issued for your license, submits CVs as PDF files, and is notified when each structured record is ready to collect as JSON. No CV data travels unencrypted, and nothing is retained beyond the terms of your license.
Can we choose what the engine extracts and verifies?
Yes. We tailor which CV data the engine extracts, and which of that data is checked against the public record, to what your platform needs.
What does a record contain?
Up to nine data classes: person and standing, education and training, employment, advising, publications, research funding, honors, research interests, and institutional identity. Classification fields use closed vocabularies, and institutions, publications, and people carry persistent public identifiers.
Which CVs does it handle?
Academic CVs in any layout and discipline, from any country. The engine understands CVs in most major languages and recognizes international institutions and venues; records are returned in English.
What does "unconfirmed" mean?
That no independent source confirmed the entry. The CV's own content is kept and plainly labeled. Unconfirmed is not the same as false: many genuine achievements are not in any public database.
Is verification a judgement about a person?
No. Verification is a data-quality check: it shows which entries an independent source corroborates. It is not a finding that anything on a CV is wrong, and it should not be used as one.
Is it the same engine as CVParsa and CVPaaS?
Yes. CVParsa for Search Committees and CVPaaS are built on the same engine. Integrations make it available inside your own platform.
How is confidential data handled?
Files are encrypted with AES-256 at rest and TLS in transit, and are never used to train third-party AI models. Retention and data handling are agreed in each license.
What does it cost?
Licensing is scoped to each partner. Write to licenses@cevianlabs.io.