The property-graph model explained from first principles: nodes, relationships, and properties; Cypher queries for real traversal problems; why index-free adjacency beats recursive SQL CTEs for deep graph walks; fraud detection, recommendations, and knowledge graphs as concrete production use cases; Neo4j clustering and causal consistency; and an honest accounting of when adding a graph database is the wrong call.
Data-Modeling
-
Graph Databases with Neo4j -
MongoDB Done Right The document database past the hype: the embed-versus-reference decision that makes or breaks a schema, denormalization that does not rot, when MongoDB genuinely beats Postgres JSONB and when it does not, the aggregation pipeline, indexing and the working-set rule, replica sets and sharding, the read/write-concern consistency knobs, and the operational footguns — unbounded arrays and the 16 MB BSON limit — that sink naive designs.
-
dbt Core in Practice: Data Modeling That Doesn't Rot A complete hands-on guide to dbt Core — project structure, sources and staging layers, testing data quality, documentation, incremental models, and running dbt in production with Airflow.