Terminology

2225 terms, tools, patterns and metrics an architect is expected to use precisely. Each one gets a short explanation of what it is, and — where it matters — what it is commonly confused with. Search filters as you type; the column headers sort.

All areas2225 Architecture Fundamentals77 Distributed Systems107 Data Architecture110 Cloud Architecture94 Networking88 API & Integration Architecture82 Reliability & Resilience75 Observability70 Performance & Capacity Engineering72 Security Architecture86 Cost Architecture & FinOps71 Business Architecture67 Architecture Communication67 Enterprise Architecture66 Legacy Modernization71 AI-Era Architecture74 Software Architecture & Engineering71 Architecture Patterns71 Architecture Decision-Making63 The Architect's Meta-Skills61 Delivery & Release Engineering66 Platform Engineering & Developer Experience70 Testing & Quality Architecture66 Data Platform Architecture69 Streaming & Real-Time Data69 Data Governance & Semantics74 Frontend & Experience Architecture66 Edge, Mobile & IoT68 Regulatory & Data Protection Architecture65 Assurance, Audit & Model Risk69

Kind Any concept801 practice736 pattern377 metric127 case-study77 tool62 protocol45

74 terms shown.

Term	Kind	Topic	What it is
Small Model Routing Model Cascade, Tiered Inference	practice	Model Selection	Sending each request to the smallest model that can handle it, escalating to a larger one only when needed.
Spotify Discover Weekly: Three Models, One Playlist Discover Weekly	case-study	ML Platform	Spotify combined collaborative filtering, natural language processing and raw audio analysis because each covers the others' blind spots.
Streaming Passthrough Gateway Token Stream Relay, SSE Passthrough Proxy	pattern	AI Gateways	Relaying a model's token stream through a shared proxy without buffering it, which makes the gateway's capacity unit concurrent long-lived connections rather than requests per second.
Subagent Context Isolation Parallel Context Windows, Context Partitioning Across Agents	concept	Multi-Agent Systems	The property that makes multi-agent systems worth their cost - each subagent explores with its own context window and returns only a condensed result, so the system reads far more than one context could hold.
Token Budget Enforcement	practice	AI Gateways	Limiting token consumption per user, tenant, feature or time window at a central point, so cost cannot run away unobserved.
Token Cost Attribution	practice	AI Cost Management	Assigning inference spend to features, tenants and users, so that cost can be managed by the people who influence it.
Tool Authorisation Boundary Model as Untrusted Proposer, Authorise Outside the Model	concept	Prompt Injection Defence	Authorising every tool invocation against the initiating user's own permissions, outside the model - because the model cannot distinguish instructions from data and no prompt-level defence is reliable.
Tool Calling Function Calling	pattern	AI-Era Architecture	Giving a model a set of typed function definitions it can request to invoke, with the application executing the call and returning the result.
Tool Result Budget Tool Output Capping, Result Truncation Contract	pattern	Tool Calling	A hard cap on the tokens any single tool may return into the model's context, with pagination and summarisation behind it, so that one unlucky query cannot fill the window and end the run.
Tool Schema Design	practice	Tool Calling	Defining the tools available to a model — names, descriptions, parameters and errors — in a way that makes correct selection likely.
Training-Serving Skew Online-Offline Skew, Feature Skew	concept	ML Platform	A divergence between the features a model was trained on and the features computed at serving time - producing a model that performs well offline and worse in production, with nothing erroring.
Two-Stage Retrieval	pattern	Reranking	Retrieving a broad candidate set cheaply and then reordering it with an expensive, more accurate model.
Vector Database	tool	AI-Era Architecture	A store optimised for approximate nearest-neighbour search over high-dimensional embeddings.
Vector Index Quantisation Product Quantisation, Scalar Quantisation, Vector Compression	pattern	Vector Databases	Storing embeddings in fewer bits so a large index fits in memory, trading a measurable amount of recall for a multiple-times reduction in RAM and a rescoring stage to win most of it back.