evidoria

← Back to browse

Good practice

VDAB — Flemish Public Employment Service AI for job matching and employability assessment

Belgium · Brussels · See the Belgium profile

VDAB has deployed multiple AI systems since 2018: a multilingual deep-learning job-matching engine; Orient 2.0 (orientation test reduced from 45 to 10 min, 10,000 users in 2 weeks); and a 2024 employability model assessing all 213,429 registered Flemish jobseekers.

45 to 10 minutes
Orient 2.0 completion time
114 to 58
Orient 2.0 questions
10,000
Orient 2.0 users (first 2 weeks)
213,429
Jobseekers assessed for employability (Aug-Sep 2024)
18%
Jobseekers with >50% employment probability (2024)
56%
Jobseekers with <35% employment probability (2024)
VDAB — Flemish Public Employment Service AI for job matching and employability assessment

Details

Maturity
Established
Promoter
VDAB (Vlaamse Dienst voor Arbeidsbemiddeling en Beroepsopleiding)
Period
2018-present
Keywords
employment service, job matching, employability prediction, deep learning, career orientation, public employment

Context

VDAB, the Flemish Public Employment Service serving about 1.8 million jobseekers, has progressively deployed AI since 2018: a deep-learning job-matching engine, the Jobbereik occupation-visualisation tool, an ML-personalised Orient 2.0 orientation assessment, and a 2024 employability-probability model covering all registered jobseekers.

Results

Orient 2.0 cut assessment completion time from 45 to 10 minutes and questions from 114 to 58, attracting 10,000 users within two weeks of launch; the August-September 2024 employability model assessed all 213,429 registered Flemish jobseekers, finding 18% with a >50% employment probability and 56% with a <35% probability. No published algorithmic audit or independent evaluation links AI use to actual employment outcomes.

Implementation

Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful