The Markko Liutkevičius thesis „Enhancing Public Employment Services with AI-enabled Virtual Competence Assistant“ explores how AI can transform Public Employment Services (PES) to address labor market challenges caused by rapid technological changes and external factors like the COVID-19 pandemic and geopolitical conflicts. With traditional skills becoming outdated and new skills in demand, the EU has set ambitious targets to ensure a skilled, adaptable workforce. Moreover, labor shortages and demographic trends—such as an aging workforce—require urgent action, especially for underrepresented groups. The research focuses on AI-enabled solutions to improve PES, exploring the current situation in the Estonian PES as an example labor market service provider in the EU.

Two critical problems are identified: first, the lack of intelligent matching between citizens’ CVs and job advertisements; second, the failure to incorporate skills in job matching. Currently, Estonian citizens must manually search job listings, and there are no modern solutions to assist them in finding appropriate job or training opportunities based on their skills. This leads to inefficiencies for both job seekers and employers. To address these challenges, the thesis proposes developing a Virtual Competence Assistant (VCA), an AI-powered service designed to enhance job matching and skills development while incorporating local labor market intelligence (LMI). The initial scope of the VCA is on matching citizens' skills with job opportunities, as job recommendation systems are more developed in existing research compared to training recommendation systems or local LMI.

A key aspect of the research is integrating the European Classification of Skills/Competences, Qualifications, and Occupations (ESCO) into AI-enabled services to match job seekers with opportunities. The thesis stresses the importance of continuous, iterative development of AI services in PES, focusing on user feedback and ethical deployment. Future research should explore training recommendation systems, local LMI, and the use of AI in PES across multiple languages and regions. The goal is to create a more efficient, inclusive, and dynamic labor market by matching not only the unemployed but all citizens with jobs and training opportunities, utilizing insights from the local labor market, and required skills.

The thesis „Enhancing Public Employment Services with AI-enabled Virtual Competence Assistant“ is published in the Digital Collection of TalTech Library.

Supervisors:

• Prof. Sadok Ben Yahia, TalTech, Estonia;
• co-supervisor: Prof. Innar Liiv, TalTech, Estonia.

Oponents:

• Dr. Marisa Ponti, University of Gothenburg, Sweden;
• Prof. Ricardo Colomo-Palacios, Universidad Politécnica de Madrid, Spain.

Follow public defence in Zoom

Meeting ID: 920 7436 7881
Passcode: 871344