Data Engineering and Architecture - 6-Month

Luxembourg National Data Service (LNDS) is a brand of PNED G.I.E. an organisation created by the Luxembourg Government, to implement Luxembourg’s strategies in research, innovation, and digitalisation. LNDS enables value creation from secondary use of data, for public and private partners and supports the sharing and re-use of public sector data, in a trustable manner. The LNDS service portfolio includes know-how, capabilities, tools, infrastructure, and data services. Through efficient & responsible use of data and improving the secondary use of data, LNDS will support the acceleration of economic, ecological, and societal transitions.

Purpose of the job / project

You will support an AI-plugged data space ecosystem project.

Context and motivation: the Data Space Support Centre (DSSC) coordinates the development of European data spaces and defines which standards a data space should use for interoperability, trust, and governance.

This project intends to enable semantic and conceptual interoperability jobs by delegating to an LLM the task of checking for mappings of meanings and concepts, aligning semantic discrepancies framed as proposals or assertions before the final human review, and accepting a new, admissible data vocabulary enrichment. The AI semantic mapper automatises the search for plausible candidates across existing SKOS mappings. LLM is also involved in drafting the first-pass Data Contract.

• Understanding of metadata standards
• Familiarity with Python
• Curiosity about open data ecosystems and metadata-driven data governance
• Basic familiarity with any documentation tool (Markdown a plus)
• Interest/knowledge in software development processes, version control systems (e.g., Git)
• Eager to use new AI tools, and to test edge concepts gmail
• Strong plus: Familiarity with:
• SPARQL, or JSON-LD for metadata manipulation and validation
• Linked Data, ontologies or semantic web technologies
• Strong written and communication skills in English
• Ability to write clear, concise, well-structured content