Agent Observability
Agent observability captures traces, logs, and cost data per step - so teams can debug failures and track token spend in production.
Calljmp
Glossary
A comprehensive guide to terminology and concepts used in AI agents, workflows, orchestration, and modern development.
A
Agentic Backend
An agentic backend is the infrastructure layer that handles execution, state, memory, and observability for AI agents running in production.
Agentic Memory
Agentic memory is the mechanism by which an AI agent stores, retrieves, and updates information across steps and sessions beyond a single context window.
Agentic RAG
Agentic RAG is a retrieval pattern where an AI agent decides what to retrieve, when, and from where - dynamically, across multiple steps. Learn how it works in production.
Agentic Runtime
An agentic runtime is the execution engine that runs AI agent code, manages step lifecycle, persists state, and handles failures in production.
Agentic Workflow
An agentic workflow is an AI process where a model reasons, calls tools, and iterates toward a goal. Clear definition, mental model, and how it differs from traditional workflows.
Agents as Code
Agents as code means defining AI agent behavior entirely in source code - versioned, tested, and deployed like any other software.
AI Agent Evals
AI agent evals are structured tests that measure whether an AI agent behaves correctly, consistently, and safely across defined inputs and scenarios.
AI Agent Orchestration
AI agent orchestration coordinates how multiple agents are triggered, sequenced, and connected to complete a shared goal.
C
Context Engineering
Context engineering means designing what an AI agent sees at each step - what to include, exclude, and how to structure it for accurate outputs.
D
Durable Execution
Durable execution is a runtime guarantee that a long-running process survives crashes, timeouts, and restarts by checkpointing state at every step.
H
Human-in-the-Loop (HITL)
Human-in-the-loop (HITL) is a pattern where an AI agent pauses execution and waits for a human decision before continuing.
L
LLM Guardrails
LLM guardrails are constraints that control what an AI agent can output or act on - blocking unsafe, off-topic, or policy-violating behavior at runtime.
M
Multi-Tenant AI Infrastructure
Multi-tenant AI infrastructure isolates agent workloads per user on shared infrastructure - no data leakage between tenants.
S
SaaS AI Copilot
A SaaS AI copilot is an AI agent embedded in a SaaS product that assists users with tasks in context - without leaving the application.
Stateful AI Backend
A stateful AI backend persists execution state across steps, sessions, and failures - so AI agents complete long-running tasks reliably.