Transform Your Talent Acquisition With Multilingual CV/Resume Parsing

Contact us

Multilingual resume/CV parsing offers applicant data capture that is fast and reliable

Best-in-class multilingual resume/CV parsing available for 23 languages to easily capture candidate data and transform into structured data

Textkernel’s highly accurate CV/resume parser, Extract!, supports recruiting organizations around the world to effectively and efficiently process large volumes of candidate documents. Transform the millions of candidate applications into structured data that can be used to filter, search and rank candidates.

Tested by prospects and clients against top competition with class-leading results

Thanks to the latest Natural Language Processing (NLP) technology, time-consuming and potentially error-prone manual data entry is a thing of the past.

*testing performed in 2020, results rank from worst (dark red) to best (dark green)

Key benefits of the Textkernel CV/Resume Parsing solution:

Dramatically less time required to process and shortlist candidates without compromising on result accuracy.

Our AI technology allows your recruiting team to focus on building human connections. The one thing that AI can never replace.

Our customers demand the highest quality results, otherwise the time savings gained through automation would be of little value. Textkernel continuously explores new techniques to optimise its extraction models.

Since 2017, Extract! has been powered by deep learning which has increased the parsing accuracy of even the most challenging cv formats. Learn more about how Textkernel was the first to launch Deep Learning to improve our parsing software quality.

Register to get updates on our cv/resume parsing improvements.

Increase candidate conversion by incorporating our parsing at the very beginning of the candidate journey.

Our Extract! technology is not just a backend process. We have developed the ability to embed our parsing software into career portals and job sites. The benefits? Dramatically quicker and easier candidate application process that provides an improved candidate experience.

This improves your candidate experience while also giving your recruitment teams the ability to tailor the candidate journey based on their skills.
Learn more.

The best multilingual resume parsing quality across 23 languages

Chinese parsing
Croatian parsing Czech parsing Danish parsing
Dutch parsing English parsing English parsing Finnish parsing
French parsing German parsing Greek parsing Hungarian parsing
Italian parsing Japanese parsing Norwegian parsing Polish parsing
Portuguese parsing Romanian parsing Russian parsing Slovak parsing
Slovenian parsing Spanish parsing Swedish parsing Turkish parsing

Discover how our parsing is the basis for both matching people and jobs and reliable business and market intelligence that can help you:

Spend dramatically less time processing and shortlisting candidates without compromising on result accuracy

Increase candidate conversion by incorporating our parsing at the very beginning of the candidate journey

Expand across borders with the highest accuracy across 23 languages

Expand internal projects with our Skills Normalization Taxonomy of more than 135,000 terms that describe 11,000+ skills


We work with the best


“We are experiencing 98% parsing accuracy across all types of job application submissions. When you parse several millions resumes a year, you need the reliability of an established provider like Textkernel.” – Jerome Ternynck, CEO, SmartRecruiters

USG People

“Textkernel’s semantic search provides consultants not only with candidates they actually thought about, but also with useful suggestions of other suitable candidates. 75% of our consultants rate Textkernel’s interface as very user-friendly, which made an extensive training obsolete.” – Ton Sluiter, Marketing Manager at USG People

Learn more


Textkernel has integrated its business development and sourcing tools with Bullhorn to enhance enterprise recruiter productivity

Learn more


Stay informed about webinars, labor market analysis and AI knowledge sharing.

Subscribe now


Follow us on LinkedIn to stay connected to the Textkernel world.

Follow us on LinkedIn
Discover Extract

Extract! resume parsing features

Textkernel’s extremely accurate resume parsing software, Extract!, is constantly being fine-tuned and improved by our large team of research engineers. Read more about the features of this product below.

How does Resume or CV Parsing work?

The resume or CV is imported into parsing software and the information contained within the document file is extracted and distilled into its elements so that the data can be categorized according to predefined fields. 

The extracted data is then automatically normalized. This means it is categorised according to a standard or customer specific format. Normalization ensures better searchability and analysis of the data processed.  

Textkernel also offers her customers the ability to enhance their parsing offering with normalisation to a specific or even custom standard.

The O*NET Profession Classification– The O*Net professional code contains hundreds of standardized and occupation-specific descriptors of approximately 1,000 occupations covering the entire U.S. economy.

ISCO profession Classification – The International Standard Classification of Occupations (ISCO) is one of the main international classifications for which the International Labor Organization is responsible.

Textkernel Profession Classification – A classification including over 4,200 professions curated by Textkernel over the past 10 years thanks to access to over one billion job vacancies.

Textkernel Skills Classification – A classification carefully built and curated by Textkernel R&D, based on the analysis of millions of candidate documents and job vacancies processed.  The Textkernel Skills Normalization Taxonomy currently contains about 135,000 terms that describe just over 11,000 skills, which are divided over four categories: 

  1. Professional skills
  2. IT skills
  3. Soft skills
  4. Languages

Learn more about how Textkernel’s Skills Classification can benefit your organization.

Once the resume has been parsed, a recruiter can easily search their database for search terms required to generate a shortlist of relevant candidates. The Textkernel parsing software is essential for powering semantic search.  Semantic search is a powerful search technology that adds context to the search terms and tries to understand intent in order to make the results more reliable and comprehensive. Now you can ensure that your recruiters don’t overlook potentially relevant candidates that might have otherwise been overlooked.  Learn more about Textkernel’s Search! offering.

How does Machine Learning and Natural Language Processing power Textkernel’s CV or Resume Parsing?

Machine learning is the technology that enables resume parsing.  Thanks to hundreds of hours spent by human annotators across all our languages, large volumes of cvs are broken down into their component parts:  personal and/or contact information, education, work experience, languages, etc.  Then our algorithms are fed millions of cvs to ‘train’ and reinforce the patterns already deciphered by the human annotators.

Looking for more information on how it works?

Sequence Labelling in Deep Learning (Or The Magic Behind Extract 4.0)

Or visit our newsroom.  Alternatively, sign up for our newsletter to stay informed of our latest parsing advancements.

What about the data security of parsed information?

Textkernel takes the security and privacy of our customers’ data very seriously. Our stringent data security procedures ensure our customers can be confident that we are handling their data assets with utmost care and consideration.    Learn more from our Information Security Officer.

Our Integrations

Click here to find out more about how this works for...



Schedule demo

Schedule a demo with us

Discover how AI-powered technology can work for your business.