Context Engineering
An engineering practice that optimizes AI model output by proactively supplying and organizing contextual information, such as project specifications, domain knowledge and tool configuration. It is a key method for improving the effectiveness of AI coding assistants.
Core Facts
AI development has evolved through four stages: single large model calls, prompt engineering, agents and tool calls, and loop engineering.
Context Engineering is a new paradigm that provides AI with a complete, systematic context, including project documentation, code examples, coding rules, development patterns, and validation mechanisms.
大多数 AI 编程智能体无法有效完成复杂任务的主要原因是上下文缺失，而非模型能力不足
Timeline (last 90 days)
Context Engineering is understanding the resource limits of an AI model's context window and properly managing external data sources such as code, tools, skills, and MCP servers.
Context engineering is good at providing information, but not at continuously exerting control during execution; when a task requires multiple steps, multiple rounds of error correction and dynamic strategy adjustment, merely providing materials cannot guarantee the AI stays on track.
The practical advice is to give the model less, cleaner, more relevant context, and to explain what the terminology means.
Providing design files as context is a common industry practice for LLM applications, called context engineering or few-shot example guidance, and the principle is that the model is influenced by existing patterns in the input.
Prompt caching, memory, compression, and retrieval are four mechanisms that are often confused but solve different problems: prompt caching reuses input processing, memory preserves information across interactions, compression reduces the context carried, and retrieval selects relevant information.
An analogy can be used to understand it: the LLM is the CPU responsible for reasoning, the context window is the RAM serving as working memory, and context engineering is the operating system that decides what gets loaded, when it gets loaded, and in what format.
Issues of reliability, security, cost, and platform fragmentation can ultimately all be traced back to context.
Andrej Karpathy described context engineering as 'the delicate art and science of filling the context window with just the right information to serve the next step.'
Context engineering is the natural extension and upgrade of prompt engineering, with prompts being only one part of context.
Placing key instructions in prominent positions, using clear separators to distinguish content types, and annotating retrieved materials with their sources can help the model understand and use context more accurately.
39 more timeline events
All Facts (20)
AI development has evolved through four stages: single large model calls, prompt engineering, agents and tool calls, and loop engineering.
90%VerifiedContext Engineering is a new paradigm that provides AI with a complete, systematic context, including project documentation, code examples, coding rules, development patterns, and validation mechanisms.
80%Verified大多数 AI 编程智能体无法有效完成复杂任务的主要原因是上下文缺失，而非模型能力不足
80%VerifiedThe system prompt is the highest-priority instruction layer in large language model conversation architecture, injected before user messages and influencing how the model processes all subsequent inputs.
65%VerifiedContext engineering refers to the engineering practice of systematically designing, managing, and optimizing the contextual information fed into large language models, which differs from prompt engineering in that the latter focuses on wording optimization for a single interaction.
65%Unverified吴恩达认为过去几年行业注意力集中在提示工程和上下文工程上，但对于需要持续运行数天甚至数周的Agent来说这远远不够
90%Unverified在真实上下文工程场景中，上下文散布在多个来源：本地文件系统、数据库、Web、长期记忆等，每种上下文源需要对应的搜索工具
90%UnverifiedAGENTS.md的实践与'上下文工程'（Context Engineering）的理念密切相关，代表了从提示工程到上下文工程的演进
85%Unverified上下文工程的重心已从'如何塞进更多信息'转变为'如何确保正确的信息出现在正确的位置'
80%Unverified在AI编程中，提前设计好数据结构并提供JSON文档给AI（即'上下文工程'）能大幅提升代码生成的准确度，这是区分AI编程新手与熟练用户的核心能力差异之一
75%Unverified行业正在经历从'模型中心'到'系统中心'的范式转移，单纯提升模型参数和训练数据的边际收益在递减
70%UnverifiedContext Engineering is defined as a systematic methodology that goes beyond prompt engineering, with prompt engineering being a subset of context engineering.
60%UnverifiedContext engineering is a further evolution beyond RAG, focusing not only on recall rate of information but also on the structural quality, credibility levels, and conflict marking of the injected context.
60%Unverified上下文工程使得零编程基础的用户也能开发复杂项目
60%UnverifiedContext Engineering is understanding the resource limits of an AI model's context window and properly managing external data sources such as code, tools, skills, and MCP servers.
50%UnverifiedContext engineering is good at providing information, but not at continuously exerting control during execution; when a task requires multiple steps, multiple rounds of error correction and dynamic strategy adjustment, merely providing materials cannot guarantee the AI stays on track.
50%UnverifiedThe practical advice is to give the model less, cleaner, more relevant context, and to explain what the terminology means.
50%UnverifiedProviding design files as context is a common industry practice for LLM applications, called context engineering or few-shot example guidance, and the principle is that the model is influenced by existing patterns in the input.
50%UnverifiedPrompt caching, memory, compression, and retrieval are four mechanisms that are often confused but solve different problems: prompt caching reuses input processing, memory preserves information across interactions, compression reduces the context carried, and retrieval selects relevant information.
50%UnverifiedIssues of reliability, security, cost, and platform fragmentation can ultimately all be traced back to context.
50%