Remote Data Scientist for AI Agent & Knowledge Graph

adesso Belgium; United States; Remote; USD 110,000 - 180,000; Full time

Benefits offered by this job

Flexible freelance setup

Remote work with EU travel eligibility

International team support

Job summary

adesso Belgium is seeking a data scientist to design and operate a production-grade AI agent platform data foundations, including canonical models, a knowledge graph, and a semantic/taxonomy layer. You will own data products from ingestion to canonical form, with a multi-tenant, versioned architecture.

Responsibilities include building data products, mastering data governance, and implementing hybrid retrieval architectures that combine graph databases and embeddings to support AI workloads.

Qualifications

• Experience taking a knowledge graph system to production with schema design and tenant isolation.
• Strong background in semantic modeling and ontologies, taxonomies, glossaries.

Responsibilities

• Build data products and canonical modeling end to end (Supplier Master, Spend Cube, Contract Register).
• Stand up knowledge graph, ontology, and tenant separation; manage graph-plus-vector retrieval design.
• Develop master-data pipelines, matching rules, golden records, and governance workflows.
• Create extraction pipelines from sources like SharePoint with confidence scoring and human review.
• Implement connectors to onboard new data sources without core changes to graph/service.
• Establish data quality feedback loops to improve data products over time.
• Work in GitHub/Azure with PRs, tests, and CI/CD; version schemas and configs.

Skills

Knowledge graph production

Semantic modeling

Graph databases

Python

Data product development

Hybrid retrieval architectures

CI/CD

SharePoint data extraction

Tools

Neo4j

SPARQL

RDF/OWL

GitHub

Azure DevOps / CI

Job description

adesso Belgium is seeking a data scientist to design and operate a production-grade AI agent platform data foundations, including canonical models, a knowledge graph, and a semantic/taxonomy layer. You will own data products from ingestion to canonical form, with a multi-tenant, versioned architecture.

Responsibilities include building data products, mastering data governance, and implementing hybrid retrieval architectures that combine graph databases and embeddings to support AI workloads.