Search the practice set

126 questions, 454 terms and 400 topics in 20 areas.

60 results for “AI Observability”

Terminology · 20
term

AI Gateway

A shared proxy in front of model providers that centralises routing, keys, quotas, caching, logging and safety policy.

AI-Era Architecture
term

Observability

The property of being able to answer new questions about a system's internal state from its external outputs, without shipping new code.

Observability
term

Alert Fatigue

The desensitisation that follows from alerts that are frequent, non-actionable, or not tied to user impact — after which real alerts are missed too.

Observability
term

Application Performance Monitoring

Instrumentation inside the application that attributes latency and errors to specific code paths, queries and dependencies.

Observability
term

Cardinality

The number of distinct time series produced by a metric, which is the product of the distinct values of all its labels — and the main driver of monitoring cost.

Observability
term

Context Window

The maximum number of tokens a model can attend to in one request, holding the system prompt, history, retrieved context, tools and the answer.

AI-Era Architecture
term

Correlation ID

A single identifier attached to one logical operation and included in every log line it produces, anywhere in the system.

Observability
term

Distributed Tracing

Following one logical request across every service it touches by propagating a shared trace identifier and recording timed spans.

Observability
term

Embedding

A dense numeric vector representing a piece of content, positioned so that semantically similar content sits nearby.

AI-Era Architecture
term

Golden Signals

The four measurements that cover most of what matters for a request-driven service: latency, traffic, errors and saturation.

Observability
term

Guardrail

A deterministic check applied to a model's input or output, enforcing rules that cannot be left to the model itself.

AI-Era Architecture
term

Health Check

An endpoint the platform polls to decide whether an instance should be restarted or should receive traffic — two different questions needing two different checks.

Observability
term

Human in the Loop

Requiring human review or approval at a defined point in an automated flow, chosen by the reversibility and cost of the action.

AI-Era Architecture
term

LLM Evaluation

A repeatable measurement of whether an AI system's outputs are good enough, on cases that reflect the actual task.

AI-Era Architecture
term

Model Context Protocol

An open protocol that standardises how AI applications connect to external tools, data sources and prompts.

AI-Era Architecture
term

Model Router

Directing each request to a model chosen by the task's difficulty, cost and latency budget, rather than sending everything to the largest model available.

AI-Era Architecture
term

Pipeline Orchestration

Coordinating the execution of data tasks by dependency rather than by clock, with retries, backfill and observability built in.

ETL & ELT
term

Prompt Injection

An attack in which text from an untrusted source is interpreted by the model as instructions rather than as data.

AI-Era Architecture
term

Prompt Registry

A versioned store of production prompts with their model bindings, parameters and evaluation results, so a prompt change is a reviewable, traceable, reversible deployment.

AI-Era Architecture
term

Prompt Versioning

Treating prompts as versioned, reviewed, tested artefacts rather than as strings edited in place.

AI-Era Architecture
Topics · 34
topic

AI Observability

Logging prompts, versions, retrieved context and cost per request.

AI-Era Architecture — no content yet
topic

AI Cost Management

Token accounting, routing, caching and the context-window budget.

AI-Era Architecture — no content yet
topic

AI Gateways

Centralised routing, keys, quotas, caching, logging and safety policy.

AI-Era Architecture — no content yet
topic

AI-Era Architecture

General material on architecting systems that include models.

31 items
topic

Observability

General material on understanding a system from its outputs.

27 items
topic

Observability Cost

Telemetry bills, cardinality control and retention tiering.

Cost Architecture & FinOps — no content yet
topic

Agent Architectures

Loops, planning, memory and the boundaries an agent must not cross.

AI-Era Architecture — no content yet
topic

Alert Fatigue

How noise makes the real page invisible, and the structural fix.

Observability — no content yet
topic

Alerting

Symptom-based, actionable, user-impacting — and linked to a runbook.

Observability — no content yet
topic

Application Performance Monitoring

Attributing latency to code paths, queries and dependencies.

Observability — no content yet
topic

Business Metrics

Orders per minute alongside error rate, because healthy is not enough.

Observability — no content yet
topic

Cardinality

The label that multiplies series count and the bill with it.

Observability — no content yet
topic

Chunking & Retrieval

Structure-aware splitting, hybrid search and why chunking dominates quality.

AI-Era Architecture — no content yet
topic

Correlation IDs

One identifier propagated through every hop and every log line.

Observability — no content yet
topic

Dashboards

Answering 'is it us' in under a minute, for someone who was asleep.

Observability — no content yet
topic

Debugging Distributed Systems

Localising a regression when every service reports healthy.

Observability — no content yet
topic

Distributed Tracing

Reconstructing one request's path across every service it touched.

Observability — no content yet
topic

Embeddings

Dense representations, model coupling and the migration they imply.

AI-Era Architecture — no content yet
topic

Guardrails

Deterministic checks on input and output that fail closed.

AI-Era Architecture — no content yet
topic

Health Checks

Liveness versus readiness, and the check that causes the outage.

Observability — no content yet
topic

Human in the Loop

Gating by reversibility and blast radius, and avoiding approval fatigue.

AI-Era Architecture — no content yet
topic

LLM Application Architecture

The shape of a production system with a model in the request path.

AI-Era Architecture — no content yet
topic

LLM Evaluation

Held-out sets, rubric judging, CI gates and production sampling.

AI-Era Architecture — no content yet
topic

Log Management

Aggregation, retention tiering, search and the cost of keeping everything.

Observability — no content yet
topic

Logging

What to log, at what level, and what must never appear in a log.

Observability — no content yet
topic

ML Platform

Feature stores, training pipelines, registries and deployment.

AI-Era Architecture — no content yet
topic

Metrics

Counters, gauges and histograms, and percentiles rather than means.

Observability — no content yet
topic

Model Selection

Capability, latency, cost and the evaluation that decides between them.

AI-Era Architecture — no content yet
topic

Multi-Agent Systems

Coordination, hand-off and whether more agents actually help.

AI-Era Architecture — no content yet
topic

OpenTelemetry

Instrumenting once against an open standard rather than a vendor agent.

Observability — no content yet
topic

Profiling

Continuous CPU and memory attribution in production.

Observability — no content yet
topic

Prompt & Version Management

Prompts as reviewed, versioned, evaluated production configuration.

AI-Era Architecture — no content yet
topic

Prompt Injection Defence

Breaking the private-data, untrusted-input, outbound-channel combination.

AI-Era Architecture — no content yet
topic

RAG Architecture

Retrieval, grounding, citation and the permissions RAG can enforce.

AI-Era Architecture — no content yet