Decoder. plain-English AI glossary

Containerization

● Core

Packaging model, dependencies, and runtime into a lightweight, portable unit that runs anywhere.

Think of it like

Shipping goods in a standardized cargo container that fits any port and truck.

Example

A Dockerfile that installs dependencies, loads model weights, and runs an inference server—deploy to cloud or on-prem unchanged.

How it actually works

Container includes OS libraries, Python environment, model, and app. Isolates dependencies and avoids "works on my machine" problems. OCI standard ensures compatibility. Overhead is minimal compared to VMs but requires orchestration at scale.

For product teams

Simplifies deployment across environments and enables faster, more reliable scaling.

For engineers

Image built from Dockerfile, pushed to registry, pulled and run by orchestrator; filesystem and process isolation guarantee consistent runtime.

Kubernetes — Enabler for.

Read anything AI without the jargon

Look up any term in plain English, or save terms as you read with the free Chrome extension.

Open Decoder Add to Chrome