Technical / dev audience · 9 keyword pages
Agent infrastructure is the foundational layer of compute, storage, networking, and runtime services that agents need to operate reliably at scale. This includes execution environments (where agent code runs), state management (how agents remember context acro...
AI agent frameworks are developer tools and libraries that provide the scaffolding for building, testing, and deploying AI agents. Examples include LangChain, CrewAI, AutoGen, and custom orchestration layers. They typically handle the agent loop (observe → thi...
MCP is an open protocol (originated by Anthropic) that standardises how AI models connect to external data sources and tools. Think of it as USB-C for AI — a universal plug that lets any model talk to any tool without custom integration code for each pair. MCP...
Tool calling architecture defines how an AI agent discovers, selects, invokes, and processes results from external tools and APIs. It covers the full lifecycle: tool description formats (so the agent knows what's available), parameter schema validation (so cal...
Context engineering is the discipline of carefully constructing, compressing, and managing the information fed into an LLM's context window to maximise the quality of its outputs. It's more than prompt engineering — it includes retrieval strategies (what to pu...
Agent memory systems manage how AI agents store, retrieve, and use information across interactions and sessions. This includes short-term memory (current conversation context), working memory (task-relevant information assembled for a specific job), long-term...
Agent pipelines are structured sequences of operations that an AI agent executes to complete a task. Unlike simple prompt → response flows, pipelines chain multiple steps: data retrieval, analysis, tool calls, validation checks, human approval gates, and outpu...
LLM orchestration is the practice of coordinating multiple LLM calls, tool invocations, and data flows to accomplish complex tasks. It covers routing (which model handles which sub-task), sequencing (what order things happen), parallelisation (what can run sim...
Developer-first AI tooling refers to products and platforms designed primarily for developers building AI applications, rather than end-users consuming them. This includes SDKs, CLI tools, local development environments, testing frameworks, and deployment pipe...