Skip to content

Analyse

Automated CV Analysis for Recruitment

Built an end-to-end CV screening tool — OCR, parsing, keyword extraction, automated job matching, candidate scoring and recruiter interface — cutting review time from 40 seconds to 3 seconds per CV. Winning project (NDA).

WinnerOCRNLPCV ParsingMatching & ScoringRecruitmentAutomationNDA

The challenge

Carried out during a three-week Business Deep Dive for Grandir Group, the challenge addressed high-volume recruitment, where screening thousands of applications manually took around 40 seconds per CV. Further operational details are confidential under NDA.

The question

How can thousands of unstructured CVs be automatically parsed, matched and scored against job requirements to cut screening time from 40 seconds to a few seconds per candidate?

What I did

Built an automated pipeline combining OCR and document parsing to extract structured candidate profiles, keyword detection, and a matching and scoring algorithm aligned with recruitment criteria, surfaced through a dedicated user interface.

Backstage · Data sources

Thousands of candidate CVs in varied formats along with job requirement criteria (specific datasets remain confidential under NDA).

The result

The team won the Business Deep Dive. The solution reduced processing time to 3 seconds per CV compared with 40 seconds previously — a 92% reduction in initial screening time.