CV/Resume Parser - Release Notes Archive
Looking for current release notes?
View current release notes on the Bullhorn Hub
November 25, 2025
Enhanced
- Improved recognition of French driver's licenses
- Fixed an issue with non-ASCII characters in email prefixes
November 20, 2025
Enhanced
- Fixed an issue with non-ASCII characters in email prefixes
November 13, 2025
Enhanced
- Improved normalization of "staff physician" jobs
- Various IT networking skills added to the skills taxonomy
- Various improvements to French profession normalization
- Various additions to the French profession taxonomy
November 12, 2025
Enhanced
- Improved recognition of British phone numbers
October 20, 2025
Enhanced
- Improved recognition of Master degrees in CV and vacancy parsing (no longer misclassifying terms like 'Scrum Master')
October 14, 2025
Enhanced
- Improved character recognition from PDF documents.
- Improved Pdf-to-text conversion for all languages
September 30, 2025
Enhanced
- Improved localization for ADR Certification.
- IT skill coverage improvements in all languages
- French skill normalization improvements
- Improved taxonomy coverage of HR job titles and skills in English
- Improve coverage of modern IT skills in skill extraction and normalization.
- Improved normalization of German assembly supervisor jobs
- Various French job titles added to the professions taxonomy
- Skill normalization added in 5 new languages: Bulgarian (BG), Estonian (ET), Latvian (LV), Korean (KO) and Thai (TH)
September 1, 2025
Enhanced
- Improved US state classification: Washington state vs Washington D.C.
August 4, 2025
Enhanced
- Improved name extraction from German CVs
- Improved parsing of multi-line LinkedIn URLs in CVs
June 30, 2025
Enhanced
- Various truck driven positions added to the profession taxonomy
- Improved normalization of German pilot professions
- Cloud-related skills added to the skills taxonomy
- 10 new skills added to the skills taxonomy
Fixed
- Fixed wrong normalization of German 'kochen'
- Fixed an issue concerning the recognition of .NET skills
May 26, 2025
Fixed
- Fix for Canadian addresses being extracted as french names.
April 14, 2025
Enhanced
- Improved name parsing in case a middle name is present
April 9, 2025
Enhanced
- Improved Chinese job title normalization (simplified and traditional)
March 31, 2025
Enhanced
- Improved Japanese skill extraction
- Better coverage of English certifications
- Added various Dutch Certifications to the certifications taxonomy
- Improved normalization of "Customs and Trade Associate"
- Added profession 'Surface treatment operator' to the professions taxonomy
- Fixed wrong normalization of 'Commercial employee' jobs
- Fixed normalization of branch manager jobs
- Improved normalization of Optician jobs in German
March 24, 2025
Enhanced
- Improved language level mapping in French
- Improved contextual validation of skills
March 3, 2025
Enhanced
- Fixed an issue with missing address info from fixed-format CVs
- Improved name extraction in English and Dutch
- Improved handling of column CVs with sidebar
February 18, 2025
Fixed
- Fixed mixup of location and employer's name in German CVs
February 3, 2025
Enhanced
- Improved handling of salary information coming from structured input
January 21, 2025
Enhanced
- Improved detection of French addresses (#28225)
- Improved detection of addresses from Malaysia (#26788)
- Improved detection of addresses from Ireland (#26537)
- Improve city detection for educational institutions (#23041)
- Romanian CVs: Improve extraction of CVs from BestJobs
- Remove the country code when mentioned in the postal codes for Austria, Germany and Switzerland (#26395)
Fixed
- Fixed an issue resulting in 'Compétence' tagged as name
January 6, 2025
Enhanced
- Improved classification of BI Developer jobs
- Improve coverage of English and Hebrew skills
December 19, 2024
Enhanced
- Swiss German recognized as separate language code
- Improved classification of BI Developer jobs
- Improved classification of medical sales professions in German
- Added 4 new professions to the professions taxonomy
- Various improvements to profession normalization in Dutch, French, German, and Spanish.
- Improved normalization of electrician jobs
- Improved Swiss-French medical profession names
- Various enrichments of the English professions taxonomy
December 16, 2024
Enhanced
- Fixed wrong normalization of Dutch education level "MBO"
December 13, 2024
Enhanced
- Improve coverage of English and Hebrew skills
- Greatly improved Japanese skills extraction
- Added various US certifications to the certifications taxonomy
December 9, 2024
Enhanced
- The parser ignores hidden sample texts from Canva CVs
Fixed
- Fixed issue concerning phone numbers being parsed from URLs
November 26, 2024
Enhanced
- Improved parsing of Canadian addresses
- English CV parser: 5-10% faster parsing especially for longer documents
November 12, 2024
Enhanced
- Various improvements to profession normalization in Dutch, French, German, and Spanish.
- Added various US certifications to the certifications taxonomy
November 11, 2024
Enhanced
- Improved Dutch name parsing
- English: 10-20% faster parsing, even faster for longer CVs
October 14, 2024
Enhanced
- PDFs: improved language guessing (#24320)
- Improve detection of Canadian addresses (#25451)
October 2, 2024
Enhanced
- Swiss German recognized as separate language code
- Fixed normalization of biology/match teachers
- Fixed normalization of pool lifeguards and swimming instructors in French
- Improved normalization of 'product designer' titles
- Improved normalization of mechanical engineers in German
- Improved casing consistency of profession descriptions
- Various improvements to the French professions taxonomy
August 19, 2024
Enhanced
- Improved email extraction with special characters in the vicinity (#22903)
- Improved handling of USA education institutions (#23242)
- Improve handling of addresses written in Cyrillic (#24260)
August 15, 2024
Enhanced
- Medical Doctorates in the US no longer get classified as "Masters", but as "PhD or Professional Doctorate" (renamed from just "PhD")
August 7, 2024
Enhanced
- Improve handling of addresses from Puerto Rico (#23241)
- Improvements to country detection (#21406, #22904)
August 1, 2024
Enhanced
- Greatly improved normalization of English job titles used in non-English documents
July 24, 2024
Enhanced
- Improved French medical skill extraction
July 8, 2024
Enhanced
- Improve handling of 2 letter English first names
- Improvements to country detection (Zendesk #22394)
June 27, 2024
Enhanced
- Improved classification of job titles to the AMS profession taxonomy
- Added Golang Developer to the professions taxonomy
- Corrected normalization of Barrister jobs
- Improved normalization of Dutch 'Conducteur' job titles
- Improved normalization of (Swiss) German medical professions
- Better handling of seniority (head, C-level, etc.) of product and operations roles in profession normalization.
- Improved normalization of apprenticeship professions in (Swiss) German
- Various medical professions added to the profession taxonomy
- Improved French medical skill extraction
- Added 45 new skills based on the SkillsFuture taxonomy
June 24, 2024
Added
- New feature: option to output an HTML version of the original in which personally identifiable information is marked and can be redacted
Enhanced
- Italian CVs: improvements to address splitting (Zendesk #21262)
- Ignore emails address of certain CV generators
June 11, 2024
Enhanced
- Improved parsing of LinkedIn HTML input
- Improved parsing of UK addresses (Zendesk #2821, #3582)
May 29, 2024
Enhanced
- Added various networking technology skills to the skills taxonomy
- Added 45 new skills based on the SkillsFuture taxonomy
May 22, 2024
Enhanced
- LinkedIn HTML: better parsing for new HTML formats
- Japanese: significant improvement to many parsing fields
April 29, 2024
Enhanced
- Better handling of USA addresses in which house numbers have a dash
April 25, 2024
Enhanced
- Patch release the professions taxonomy with a fix for wrongly mapped ISCO codes
April 16, 2024
Enhanced
- Ignore emails address of certain CV generators
- Improvements to detection of Canadian addresses
April 2, 2024
Enhanced
- Improved coverage of (Swiss) German medical skills
- Added 20+ medical skills to the skills taxonomy in all languages.
- Added 10 new skills to the skills taxonomy in all languages
- Improved skill extraction of Hebrew CVs
- Various improvements to the French professions taxonomy
- Various job titles added to the professions taxonomy based on Swiss market data
- Improved Japanese Profession normalization
- Improved normalization of 'apprenticeship' professions
- Added Corporate Recruiter to the professions taxonomy
- Improved Japanese profession normalization
- Added 12 new professions to the professions taxonomy
- Added 25 new healthcare professions to the professions taxonomy
- Improved coverage of healthcare skills: added 40 skills and 275 synonyms
March 27, 2024
Enhanced
- Fixed casing of Britisch local education levels
- Improved coverage of healthcare skills: added 40 skills and 275 synonyms
March 19, 2024
Enhanced
- Fixed casing of Britisch local education levels
March 18, 2024
Enhanced
- Fixed issues with parsing of LinkedIn Public and Private HTML profiles
March 5, 2024
Enhanced
- Linkedin PDF: Better handling of names with many titles
- Improvements to handling US addresses
February 6, 2024
Enhanced
- Improvements to first and last name detection for English language CVs
- Extract national id numbers in Hebrew and South Africa CVs
January 30, 2024
Enhanced
- Added support for language skills Pedi and Sepedi
January 25, 2024
Enhanced
- Improved Japanese profession normalization
January 22, 2024
Enhanced
- Hebrew CVs: large improvements to parsing of course items
January 9, 2024
Enhanced
- Linkedin HTML private profiles: improve parsing in light of recent HTML changes
December 14, 2023
Enhanced
- Added support for language skills Pedi and Sepedi
- Added more Gen AI tools to the skills taxonomy
- Added Prodigy (annotation tool) to the skills taxonomy
- 13 skills added to the skills taxonomy for all Tier 1 languages
- Added Hebrew language skills
- Add skill COSYS to the skills taxonomy
- Improved the names of three German IT job titles
- Changed the gender tag (m/v) to (m/v/d) in Dutch profession normalization
- Improved normalization of IT professions in English
- Fixed wrong normalization of Front End Clerk jobs
- Improved taxonomy coverage on watchmaker professions
December 11, 2023
Enhanced
- Hebrew: Significant improvements to date and name parsing
- Portugese: Improved name extraction for certain corner cases
- Infer and return the county for US addresses (Data Model 2 only)
November 16, 2023
Enhanced
- Improve detection of ambiguous Canadian cities
October 30, 2023
Enhanced
- Improved address parsing from CV's in Malta
October 13, 2023
Enhanced
- Various improvements to Italian profession and skill normalization
- Improved recognition of language skills by written as iso code
- Added various CompTIA skills to the skills taxonomy
- 16 new skills added to the skills taxonomy
- Improved accuracy of finance and insurance skills
October 12, 2023
Enhanced
- Improved recognition of two-character language skills
October 10, 2023
Enhanced
- Improved recognition of language skills by written as iso code
September 28, 2023
Enhanced
- Improved classification of CFO/CFA job titles
- Various improvements to Italian profession and skill normalization
- Added 5 new professions to the professions taxonomy
- Renamed some French professions that exist both in English and French
- Improved Spanish profession normalization for planner jobs
- Improved normalization of French professions
- Various improvements to Italian profession normalization
- Fixed wrong normalization of cashier job titles
- Various improvements to Spanish job title normalization
September 26, 2023
Enhanced
- Improved recognition of Bachelor's degrees
- Fixed Québec (CA) education levels
Fixed
- Fixed an issue related to parsing Bachelor Degrees from jobs
September 20, 2023
Enhanced
- Improve national IDs extraction for Colombia
September 5, 2023
Enhanced
- Improved Japanese Skills Extraction
July 26, 2023
Enhanced
- CV-IT: Improved extraction of company and location in experience section
July 12, 2023
Enhanced
- LinkedIn PDF: Improve location extraction for multiple positions in the same company
June 29, 2023
Enhanced
- Various improvements to French job title normalization
- Improved normalization of Agile-related job titles
- Various improvements to Italian job title normalization
- Improved normalization of the Dutch job title "Instructeur serveren"
- 50 new professions added to the Professions Taxonomy
- Four new professions added to the professions taxonomy
- Various improvements to the German professions taxonomy
Fixed
- Fixed wrong German translation of "Police Surveillance Officer"
- Corrected the ISCO code mapping for Product Owner
- Fixed wrong normalization of Dutch job title "Adviseur (Digitale) Geletterdheid"
- Fixed wrong normalization of "Credit Control Officer"
June 27, 2023
Enhanced
- Improve name extraction in column CVs. Applies to documents in which the name is in the right column in large font
May 26, 2023
Enhanced
- Improved region classification in Japanese job parsing
May 22, 2023
Enhanced
- Better classification of German education level "Weiterbildung"
May 16, 2023
Enhanced
- Improved skill extraction for Czech, Slovak, Croation, Slovenian, Greek, Turkish, Romanian, Hungarian and Russian
- Improved the classification of the education level 'Weiterbildung' in Germany
- Improved classification of language level 'scolaire' in French
- Improved region recognition for Clipperton and Saint-Barthélemy
Fixed
- Fixed an issue concerning email addresses that contain white spaces.
May 3, 2023
Enhanced
- Keep original "https" in extracted social medial links
April 18, 2023
Enhanced
- Italian: Improved interpretation of dates
- Better handling of CVs where last name is only an initial
March 31, 2023
Enhanced
- Added ±10 SAP-related IT skills to the skills taxonomy
- Added ±500 new skills to the Skills taxonomy
- Added translations of two languages: Balochi and Aramaic
- Added various Portuguese skills to the skills taxonomy
- 6 New skills added to the Skill Taxonomy
March 30, 2023
Enhanced
- Added +- 80 German job title synonyms to the professions taxonomy
- Added 5 new professions to the professions taxonomy, and renamed 7 professions in French
- Various improvements to German job title cleaning
- 1170 job title synonyms added to the German Professions Taxonomy
- Various improvements to the German professions taxonomy
- Various improvements to Portuguese profession normalization
March 21, 2023
Enhanced
- Added translations of two languages: Balochi and Aramaic
February 22, 2023
Enhanced
- Better support for overseas territories
- Improved recognition of Chinese phone numbers
- Improved handling of abbreviated dates
- Improvements to country classification
January 24, 2023
Enhanced
- LinkedIn HTML: Add support for manually added phone numbers
January 11, 2023
Enhanced
- Improved cleaning of job titles from Ireland
- Improved Portuguese profession normalization
- Improved Japanese profession normalization
January 10, 2023
Enhanced
- Indeed CVs: added support for recognizing postal codes in address
- LinkedIn HTML: Fix rare case where experience is not extracted
December 15, 2022
Enhanced
- Various improvements to the French profession taxonomy
- (m/f) tags were removed from profession descriptions in English, and -man suffixes (e.g. craftsman) were replaced by gender neutral equivalents (craftsperson)
- Added SQL developer as a separate profession
- Three Javascript frameworks added to the skills taxonomy
- Added new diploma for French language
- Hundreds of new terms added to the Japanese skills taxonomy
Fixed
- Fixed classification of language teachers in Japanese
December 5, 2022
Enhanced
- Added SQL developer as a separate profession
November 30, 2022
Enhanced
- Skill extraction improved for all accounts from before 2020 (up to par with more recent accounts)
November 29, 2022
Enhanced
- Skill extraction improved for all accounts from before 2020 (up to par with more recent accounts)
October 27, 2022
Added
- Added education resources for Singapore
Enhanced
- Added new diploma for French language
October 18, 2022
Added
- Added support for fast reprocessing of stored CVs to add or update normalized skills. Updating an entire index is still work in progress.
Fixed
- Fixed classification of language teachers in Japanese
October 4, 2022
Fixed
- Fixed a bug regarding empty location fields
September 29, 2022
Enhanced
- Updated the profession taxonomy (Septemer-2022 release)
- 20 New professions added to the professions taxonomy
- Various improvements to the French professions taxonomy
- Various improvements to job title cleaning
Added
- Various small improvements to the French professions taxonomy
Fixed
- Fixed wrong classification of building inspector and covid tester
September 21, 2022
Fixed
- A bug was fixed for API integrations with Sourcebox. The bug occurred when a TMF was submitted to Sourcebox, and at the same time an empty file as a CV (or an invalid non-CV file) was also included. The fix is that we cover this case to add a "No CV" definition to the Trxml Document, which enables the TMF to be indexed in search.
August 23, 2022
Added
- Output skills origin information
Enhanced
- Improved country detection for fixed format CVs (e.g. LinkedIn, Indeed, etc)
- Improved handling of column CVs with emails in the lower part of the page
- Improved handling of Irish addresses
August 9, 2022
Enhanced
- Improved classification of addresses in Luxembourg
July 28, 2022
Added
- Added support for Traditional Chinese parsing and Taiwan locale.
July 10, 2022
Enhanced
- Added support for Arabic skills extraction (tier 3)
- Support for Hebrew skills
Fixed
- Solved an issue regarding ambiguity of 'Rust' in Dutch
- Improve O*NET mapping for home health aides
June 13, 2022
Enhanced
- Portugese: Large improvements to parsing quality by upgrading to Deep Learning
June 2, 2022
Enhanced
- LinkedIn PDF: fix rare experience parsing issues for French
May 19, 2022
Enhanced
- Improvements to address and phone fields for Spanish Colombia locale
May 14, 2022
Enhanced
- Added OCR support for Hebrew and Arabic documents
May 5, 2022
Fixed
- Fixed missing Umlauts in Austrian locations
Enhanced
- Improve handling of Norwegian addresses
- Improvements to address and name fields for Portuguese Brazil locale
April 21, 2022
Enhanced
- Removed trunk code from the output of phone numbers (in Data Model v2). Phone numbers are now output in international notation, e.g. +31204942496.
January 27, 2022
Enhanced
- This version of Sourcebox introduces a change where alert, prompts and confirmation dialogs no longer appear as browser prompts, but instead are consistent with the rest of the look and feel of the application. This does not affect their functionality, which remains the same. This is in response to Google Chrome deprecating the usage of these functionshttps://chromestatus.com/feature/5148698084376576
- Improve parsing of mixed language Chines CVs from liepin.com
- Romanian: improvements to address fields
December 7, 2021
Enhanced
- Various improvements to the German skills taxonomy
- Added various (digital) marketing skills to the skills taxonomy
- Improved skill disambiguation for multi-word terms in all languages
- Improved the names of about 200 skills in the German skill taxonomy
- Various improvements to Spanish and Italian translations of professions
Fixed
- Fixed ambiguity issue regarding German skill 'Schreiben von Berichten'
- Improved German translation of job group "Government Administrators"
November 22, 2021
Enhanced
- Improved handling of column CVs for all languages
- Improved country classification for Peru and better handling of Spanish date abbreviations
November 12, 2021
Enhanced
- Improved recognition of Mexican regions
November 8, 2021
Enhanced
- Improve splitting of Peruvian addresses
- Improve handling of Mexican addresses
- Significant improvements to Italian CV parsing (20-50% error reduction depending on field)
September 28, 2021
Enhanced
- Better alignment of language fluency levels in German and English
August 3, 2021
Enhanced
- Improved recognition and splitting of Chilean addresses
July 30, 2021
Enhanced
- Improved recognition of nationalities based on country codes
- Improved location mapping for French overseas territories
- Improved recognition of French subregions
Fixed
- Fixed an issue related to subregion recognition
June 21, 2021
Enhanced
- Improved extraction of LinkedIn profile links in case multiple are mentioned in the document
May 18, 2021
Fixed
- Fixed issues with CV upload and editing profile when using IE11 in mobility setups.
April 12, 2021
Enhanced
- Improve University Name splitting
- Linkedin PDF: improved candidate location extraction
- Linkedin PDF: Parsing improvements for fields that are line wrapped
- LinkedIn PDF and Indeed CVs: Improved address recognition
March 22, 2021
Enhanced
- Improved recognition of "Erzieher" in German profession taxonomy
March 19, 2021
Enhanced
- Improved recognition of (sub)regions in india and various other countries
- Improve address fields for locations from Ireland
March 15, 2021
Enhanced
- Improved recognition of (sub)regions in india and various other countries
March 11, 2021
Enhanced
- Improved recognition of (sub)regions in india and various other countries
Fixed
- Improved recognition of the country Austria in German documents
March 1, 2021
Enhanced
- Improved recognition of (sub)regions in india and various other countries
February 17, 2021
Added
- New option to disable automatic word wrapping for input formats that support soft wrap (e.g. Microsoft Word). PDF files do not support soft wrap. When enabled, longer lines are no longer wrapped at 80 character limit. Useful for customers that build an UI to present the parsing results
December 21, 2020
Added
- Added 25 ISCO and ONET mappings to the profession taxonomy
Enhanced
- A small fix in the Dutch Profession Taxonomy, resolving erroneous classifications of "Montagemedewerker Brilmonturen"
- Medical Affairs Manager added to profession taxonomy
- Small improvements to French profession normalization
Fixed
- Minor improvements to the American English profession taxonomy
December 7, 2020
Enhanced
- Error severity levels have been revised and updated for the Extract SOAP API, including advice on retrying strategies for better error handling. Some error IDs have been changed. For more details, refer to the Sourcebox API Reference documentation.
- Improved parsing speed and reduced timeouts for certain documents when highlighting of extracted values in the HTML rendering is enabled
November 23, 2020
Enhanced
- Improved extraction of names from LinkedIn profiles
- Improved detection of Indeed profiles so that those profiles are parsed with high accuracy
November 20, 2020
Fixed
- Non-existing city 'Nueva York' removed from autocomplete
November 9, 2020
Enhanced
- Improved country detection for Australia
- Improved splitting of names for French
- The highest education level is no longer derived from unfinished education items
October 26, 2020
Enhanced
- Various improvements to extraction of LinkedIn profiles
- Improved extraction of name from LinkedIn and Indeed profiles
September 28, 2020
Fixed
- Bullhorn users were not able to import candidates from external sources.
September 25, 2020
Added
- New feature for the Talentsoft Marketplace Match integration for the user can set a query term / tag as either Mandatory (Must-have) or Optional (Nice-to-have).
June 8, 2020
- In the Bullhorn-native integration, de-duplication was originally designed for single-tenant environments. It has been extended to work with multi-tenant orgs so that Bullhorn BusinessUsers will only see possible duplicates according to their Departments. This does not require configuration.
May 25, 2020
Added
- Added support for Taiwanese Address parsing
May 15, 2020
Added
- High accuracy parsing of kariyer.net profiles (Turkey)
- Parsing of Turkish, Slovenian and Croatian CVs. Documents in these languages can now also be used in Search. Semantic Search is available for Slovenian and Croatian.
Enhanced
- Improved country detection for French addresses containing German city names as street name
- Reduced the amount of documents going through OCR when no actual OCR was needed. As a result, parsing is faster and more accurate for these documents.
May 11, 2020
Added
- Parsing of Turkish, Slovenian and Croatian CVs. Documents in these languages can now also be used in Search. Semantic Search is available for Slovenian and Croatian.
Enhanced
- Improved country detection for French addresses containing German city names as street name
- Reduced the amount of documents going through OCR when no actual OCR was needed. As a result, parsing is faster and more accurate for these documents.
April 24, 2020
Added
- High accuracy parsing of kariyer.net profiles (Turkey)
March 30, 2020
Enhanced
- Improved extraction of phone numbers and language skills from LinkedIn PDFs
March 17, 2020
Enhanced
- Improved classification of language proficiency, in particular level B2.
March 4, 2020
Enhanced
- Improved classification of language proficiency levels C1 and C2.
January 20, 2020
Enhanced
- Improved name splitting for German and Dutch when titles like MSc, MA, BSc, BA are part of the name
January 6, 2020
Added
- Initial release of the Talentsoft Match Integration. This is a plugin allowing Talentsoft users to automatically find the best Candidates starting from a Vacancy, and find the best Vacancies starting from a Candidate.