This Week in Neo4j: Entity Extraction, MS Agent Framework, Context Engineering and more

Senior Developer Marketing Manager

January 9, 2026

3 min read

Welcome to This Week in Neo4j, your fix for news from the world of graph databases!

Happy New Year! This edition dives into techniques for structuring unstructured data with GliNER2, building graph-aware agents with Neo4j and the Microsoft Agent Framework and tuning context engineering pipelines for higher-accuracy AI.
I also want to draw your attention to the updated Neo4j Developer Center, making it easier than ever to learn Cypher, explore language drivers and build smarter graph-powered applications.

NODES AI, our global graph-and-AI event, is taking place April 15, and we’ve just announced the first featured speakers across themes like GraphRAG, Graph Memory and AI in Production – register now so you don’t miss the Road to NODES AI workshop series.

Happy Graphing,

Alexander Erdl

COMING UP!

Livestream: Unify Your Database Operations to Optimize Fleet Management on January 13, Neo4j Live: Building Graph Powered Mobile Apps with Flutter and Neo4j on January 20 & Discover Neo4j AuraDB – S03E08: Aura GraphRAG Agents on February 9
Conferences: Find us at jFokus, Stockholm on February 2-4
Meetup: Meet us in Glasgow, UK on January 14, San Francisco, US on January 27
All Neo4j Events: Webinars and More

FEATURED COMMUNITY MEMBER: Athulya Anil

Athulya Anil is a graduate student and AI researcher at UMass Amherst. She developed Agentic GraphRAG, a multi-agent framework that automatically infers schemas, constructs knowledge graphs, and adaptively routes queries between vector search and graph traversal without manual schema design.

Connect with her on LinkedIn.

Athulya is a featured speaker for NODES AI. Her talk “Agentic GraphRAG: Autonomous Knowledge Graph Construction and Adaptive Retrieval” walks through how specialised agents collaborate to extract entities and relationships, resolve conflicts, handle multi-hop queries, and select an appropriate retrieval strategy using explicit routing logic informed by empirical failure patterns.

ENTITY EXTRACTION: GliNER2: Extracting Structured Information from Text

Tomaz Bratanic introduces GliNER2, a technique for extracting structured entities and relations from unstructured text using prompt engineering and lightweight models. By defining clear extraction schemas and iterative refinement strategies, GliNER2 improves the precision of automated knowledge graph construction from narrative documents – a valuable tactic for GraphRAG and NLP pipelines.

MICROSOFT AGENT FRAMEWORK: Building Graph-Aware Agents with Neo4j and Microsoft Agent Framework

Zaid Zaim introduces the MS Agent Framework, a developer toolkit for building coordinated agent workflows powered by knowledge graphs. It enables you to define agent roles, orchestrate multi-step tasks, and share context seamlessly via a Neo4j graph, making it easier to build complex, stateful AI systems with explicit relational memory.

CONTEXT ENGINEERING: Zep/Neo4J Workshop: Context Engineering with Graphs + Contest

Will Lyon and Jack Ryan show in this workshop how to ingest a synthetic developer/agent dataset into Zep to build a Neo4j-backed knowledge graph, then tune how context (facts, entities, episodes + prompting) is retrieved and assembled to boost a coding agent’s eval score. The goal is to maximise hard-category accuracy under tight limits (2,000 characters and a 2-second retrieval budget), balancing completeness with whether the LLM actually uses the retrieved context.

DEVELOPER CENTER: Build Smarter Apps, Easier, with Graph Tools For Your Language

We’ve just updated our Developer Center – your central hub for all things graph development, featuring tutorials, API docs, quickstarts to GraphAcademy courses, and code samples. It helps you learn Cypher, explore drivers in multiple languages, and build graph-powered applications with confidence using the latest tooling and best practices.

CONTINUOUS LEARNING

GraphAcademy: Start with the most important basics and learn “Graph Data Modeling Fundamentals“
Learn on Your Schedule: Go deeper into graph technology on Neo4j’s On-Demand webinar library
Workshops: Join our virtual classrooms on GraphRAG & GenAI Fundamentals
New Webinar: Knowledge Graphs for AI and Intelligent Systems – Americas, Europe, Middle East & Africa, Asia Pacific