Novelcore (NVCR) — Explainable Agentic AI & Knowledge Systems

Novelcore operates across the full modern AI stack with a consistent focus on deployability, explainability, and human oversight. Key competencies include Knowledge Graph construction and metadata enrichment — end-to-end tooling for building structured, queryable knowledge representations from unstructured data using entity resolution, named entity recognition, and ontology-driven modelling. Graph Retrieval-Augmented Generation (G-RAG) combines graph-based knowledge representation with large language models, grounding outputs in traceable, structured knowledge for applications in question-answering, decision support, and intelligent assistants. Semantic search and information retrieval goes beyond keyword matching, leveraging semantic embeddings, query expansion, and knowledge graph integration for accurate, context-aware results across large domain-specific corpora. Multi-agent system design covers the full architecture of agentic AI systems — orchestration, agent-to-agent communication, human-in-the-loop governance patterns, and explainability at both local and system level. NLP and language understanding includes LLM fine-tuning, cross-lingual embedding models, argumentation mining, sentiment analysis, and domain-specific language model development, including validated Greek-language models (ORPHEAS, Themida, GEM-RoBERTa).