Installing Neo4j Desktop and Neo4j Aura (cloud)

Navigating the Neo4j Browser and Bloom

Connecting via Python (neo4j driver) and HTTP API

Cypher syntax fundamentals: MATCH, CREATE, MERGE, RETURN

Filtering with WHERE, ordering, and limiting results

Pattern matching — the heart of Cypher

Hands-on: modelling a small product catalogue as a graph

Adding nodes, relationships, and properties

Running your first queries

Importing CSV files with LOAD CSV

Using APOC (Awesome Procedures on Cypher) for bulk imports

Connecting Neo4j to relational databases (JDBC)

Think in relationships, not tables

Identifying entities, relationships, and properties

Common modelling patterns: linked lists, trees, bipartite graphs

Index strategies in Neo4j

Avoiding common modelling pitfalls (over-normalisation, super nodes)

Profiling and explaining queries

Translating business requirements into graph models

Worked example: modelling a pharmaceutical clinical trial domain

Worked example: modelling a financial portfolio and market relationships

Enforcing uniqueness constraints

Node key constraints

When to use strict schema vs. schema-free design

Ontology vs. schema vs. taxonomy

Classes, properties, and instances

Real-world ontologies: SNOMED CT, ChEBI, Financial Industry Business Ontology (FIBO)

Top-down vs. bottom-up ontology design

Defining class hierarchies and property ranges

Balancing expressiveness with practicality

Mapping OWL ontologies to the Labeled Property Graph model

Using the neosemantics (n10s) plugin for RDF/OWL import

Versioning and evolving your ontology over time

The three-layer model: ontology / reference data / instance data

Managing controlled vocabularies

Practical governance: who owns what layer?

Variable-length path queries

Aggregations, subqueries, and CALL {} syntax

Working with lists, maps, and UNWIND

Introduction to the Neo4j GDS plugin

Centrality algorithms: PageRank, Betweenness Centrality

Community detection: Louvain, Label Propagation

Similarity algorithms: Node Similarity, Cosine Similarity

Shortest path (Dijkstra, A*)

All paths and weighted paths

Practical use case: drug interaction pathways, supply chain routing

Native projections vs. Cypher projections

Running algorithms on projected graphs

Streaming vs. mutating vs. writing results

The limitations of LLMs: hallucination, static knowledge, lack of reasoning

How knowledge graphs ground LLMs in structured, verifiable facts

The rise of GraphRAG (Graph Retrieval-Augmented Generation)

Architecture overview: query → graph retrieval → LLM generation

Connecting Neo4j to LangChain and LlamaIndex

Writing graph-aware prompts

Hands-on: building a Q&A bot powered by a knowledge graph

Using LLMs to generate Cypher queries from natural language

Prompt engineering for reliable Cypher generation

Validation and safety guardrails

Hands-on: natural language interface to your Neo4j graph

Using NLP and LLMs to extract entities and relationships from documents

Named Entity Recognition (NER) pipelines

Populating a knowledge graph automatically from text

Hands-on: extracting a drug-disease graph from medical abstracts

How AI agents use knowledge graphs as long-term memory

Storing and retrieving agent state in Neo4j

Practical architecture patterns for agentic AI systems

Drug discovery and target identification

Clinical trial design and protocol management

Adverse event analysis and pharmacovigilance

Case study: AI agent generating eCRF specifications using a knowledge graph

Portfolio and asset relationship modelling

Fraud detection using graph patterns

Regulatory compliance and reporting

Case study: detecting insider trading networks

IT asset and dependency mapping

ITSM and incident root cause analysis

Master data management (MDM) with graphs

Case study: enterprise knowledge base for IT support automation

Common patterns across industries

Where knowledge graphs deliver the most ROI

Common failure modes and how to avoid them

Neo4j self-hosted vs. Neo4j Aura (managed cloud)

Sizing and capacity planning

High availability and clustering

Role-based access control (RBAC) in Neo4j

Securing graph data by subgraph

Audit logging and compliance considerations

Data lineage and provenance tracking

Managing ontology versions

Handling data quality and conflict resolution

Query performance monitoring

Index maintenance and database health checks

Backup and recovery strategies

Build an end-to-end knowledge graph solution for a domain of your choice

Options: pharma trial management, financial portfolio analysis, IT asset graph

Define the ontology and schema

Load and model the data

Write key Cypher queries

Integrate an LLM for natural language querying

Reviewing the completed project

How to extend your knowledge graph

Recommended resources and communities (Neo4j GraphAcademy, research papers)