A glossary, not a textbook.
Short definitions of the database and AI concepts our articles assume. Each one is a couple of paragraphs, with links to the deeper pieces.
Patterns
Postgres
Connection Pooling
A pooler fronts Postgres with a smaller set of upstream connections, multiplexing many client connections over them. Essential for serverless and high-concurrency workloads.
Read definitionJSONB
JSONB is Postgres's binary JSON column type. It stores structured data with indexing, operator support, and decent performance, blurring the line between relational and document stores.
Read definitionMVCC (Multi-Version Concurrency Control)
MVCC is Postgres's concurrency model. Every row is a chain of versions; readers don't block writers; writers don't block readers. The cost: bloat from long transactions.
Read definitionpgvector
pgvector is the Postgres extension that adds a vector data type, distance operators, and HNSW indexing. The 2026 default for production RAG inside the database you already run.
Read definitionRow-Level Security (RLS)
Row-Level Security is a Postgres feature that lets you attach per-row policies to a table. Queries automatically filter rows based on the caller's identity.
Read definition
AI
HNSW (Hierarchical Navigable Small Worlds)
HNSW is an approximate-nearest-neighbour index used by every major vector database. Great recall, fast queries, larger memory footprint than alternatives.
Read definitionRAG (Retrieval-Augmented Generation)
RAG combines a vector search step with an LLM generation step. Retrieve relevant chunks; pass them to the model as context; generate the answer. The standard pattern for grounded LLM apps in 2026.
Read definition