Instantly Parse Academic CVs

CVParsa is the automated solution for faculty search committees. Securely and quickly extract education, employment, scholarly interests, and publication data from dossiers and summarize your applicant pool in a simple spreadsheet. Spend your time evaluating candidates, not on tedious data entry.

Get Started Now →
CVParsa capybara managing the data extraction process.
Core Features

Deep, Accurate, and Secure Data Extraction

CVParsa is specifically designed to understand academic data, ensuring unparalleled accuracy in parsing complex sections. Our technology goes beyond basic name and contact information to pull out the details that matter to your committee. Stop manually copying data from dossiers, and spend more time evaluating candidates.

•
Useful Data
Person, Interests, Education, and Employment information is standard in all tiers.
•
Publication Counts
Parse all listed publications, and count first and last author papers, and publications in the venues you care most about.
•
Secure Processing
All files are kept strictly confidential on our secure server, which implements AES-256 encryption at rest.
Decision Support

Understand the Applicant Pool

See the composition of your applicant pool, at a glance. CVParsa can automatically generate a visual infographic summary, giving your committee an immediate, bird's-eye view of all candidates' backgrounds and credentials.

•
Save Time
Reduce hours of manual data entry to mere minutes.
•
Evaluate Fairly
Standardized data facilitates objective, apples-to-apples comparisons.
•
Customize Output
Receive a single, clean spreadsheet for easy committee sharing. Choose which columns are included.
CVParsa capybara analyzing the resulting data.

Ready to Save Your Committee Time?

Upload your ZIP file of CVs or full applications, and get your consolidated spreadsheet back today. Handle 1 applicant or 500, all in one go.

Process My CVs →

Flexible Processing Tiers

Basic

$3/file

  • Person, Interests, Education, and Employment information.
  • Publication extraction and metadata.
  • Infographic summary of applicant pool.
Select Basic

Plus

$5/file

  • Person, Interests, Education, and Employment information.
  • Publication extraction and metadata.
  • Infographic summary of applicant pool.
Select Plus

Premium

$7/file

  • Person, Interests, Education, and Employment information.
  • Publication extraction and metadata.
  • Infographic summary of applicant pool.
Select Premium

Standard pricing is per CV document processed. Multi-use licenses also available; contact licenses@cevianlabs.io for license inquiries.

Frequently Asked Questions

CVParsa produces a spreadsheet summarizing the individual CVs that you upload. Each input file becomes a spreadsheet row, and each row details a set of characteristics extracted from the CV.

Basic tier fields: Name, Gender (inferred), Email, Current job, Current institution, Current dept., Highest degree, PhD institution, PhD advisor, PhD field, PhD year, Interests keywords, Interests summary.

Plus tier fields: All Basic tier fields plus Publication count, 1st author pub count, last author pub count, and (optionally) up three venue-specific publication counts.

Premium tier: All Basic + Plus tier fields, plus a downloadable infographic visualization of the composition of the pool.

Yes, you can choose to omit fields from the output file. For instance, if you want to omit the Gender column, or the PhD institution, to remove gender or prestige signals from the spreadsheet. These options are entered prior to downloading the spreadsheet.

CVParsa uses a custom stack of modern AI tools to automatically process the semi-structured data in academic CVs. These tools are built on 10 years of expertise by the CVParsa team in analyzing employment, education, and scholarly activity by academic faculty, and leverage industry standard security protocols to process files quickly and maintain their confidentiality throughout.

Briefly, the process works like this:

  1. Create a .zip archive containing the stack of individual CVs or faculty dossiers in PDF format.
  2. Upload the archive to CVParsa for processing on our secure server.
  3. It takes about 5 minutes per 100 files, so get a cup of coffee or answer some email while you wait.
  4. Select which output format options you want.
  5. CVParsa notifies you when you output file is ready to download.

No. CVParsa cannot split apart PDFs. Each person you want to summarize must have their own PDF. If your pool contains 100 applicants, there should be 100 files, one for each applicant. Files can be full dossiers or applications, or just academic CVs.

No. CVParsa only works on PDF files at this time.

CVParsa can understand documents in most major languages, and recognize international degrees, institutions, and publishing venues. CVParsa always returns results in English.

CVParsa supports all academic fields and areas of study. So long as the CV is structured like an academic curriculum vitae, it should work.

If academics in your field do not publish articles or books, the publication parser will likely produce incomplete output. For instance, at this time, it does not support parsing for artistic performances.

If a publication has a standard bibliographic structure (e.g., a journal article, a peer-reviewed conference publication, a monograph, etc.), CVParsa will try to extract it. All extracted publications are checked against the Crossref database to correct errors when possible, but all extracted publications are counted.

Yes. License options for securing multiple CVParsa jobs are available. License inquiries can be directed to licenses@cevianlabs.io

Yes. In addition to the option of paying per file parsed, Cevian Labs offers site licenses to institutional partners, and supports tax exempt payments via invoice. Please contact us if you are interested in our site license options. (If using CVParsa requires Cevian Labs being an approved vendor for your university, please contact us with details about that process.)

We take the confidentiality of applicant information extremely seriously. File uploads are encrypted, our secure server uses AES-256 encryption at rest to protect uploaded files, and files are never used to train third party AI models. PDFs are deleted after CVParsa finishes, and original files are retained for only 30 days in order to handle customer support requests.

Our data retention policy is detailed in the Privacy Policy for CVParsa. Briefly, copies of some files may be retained for internal audits, service improvement efforts, and as required/permited by law.

To use CVParsa, you must agree to the End User License Agreement (EULA), which grants us permission to use the submitted files to provide you with the resulting summaries.

Academic CVs varies widely in format, styles, and conventions across fields, subfields, and even individuals. CVParsa uses advanced natural language processing algorithms, supported by extensive domain customization to identify and extract relevant information. At the same time, no automated tool is perfect, and CVParsa does make some mistakes.

In our extensive testing on 100s of CVs spanning every academic field, CVParsa is about 99% accurate. In a stack of 100 CVs, CVParsa may make a few dozen mistakes across all the pieces of information it extracts. These mistakes are sometimes in distinguishing different types of publications, or understanding sometimes unique or unusual academic titles. If you notice any parsing mistakes, please let us know, so that we can improve CVParsa's accuracy on that particular type of extrated information.