46 篇文档带有标签「Intermediate」

How AI agents store, retrieve, and reason over information across turns and sessions.

Best practices for writing system prompts that produce reliable, well-scoped AI agent behavior.

AI for perception, planning, and control in robotics.

Ensuring AI systems are robust, aligned, and safe.

Claude's native function/tool calling mechanism using JSON schema definitions, tool_use and tool_result message types, with support for multi-turn tool use, parallel calls, and streaming.

Microsoft's multi-agent conversation framework enabling LLM-powered agents to collaborate via structured message exchanges, with built-in code execution and human-in-the-loop support.

Gemini 的编码器-解码器前身；用于摘要和生成的去噪预训练。

DeepSeek AI 的开放权重大语言模型，具有强大的推理和代码能力；MoE 和高效扩展。

阿里巴巴的大语言模型家族；多语言、编程和长上下文支持。

Step-by-step reasoning to improve LLM outputs.

Continuous integration and delivery adapted for machine learning — testing data, models, and code together.

扩展 Claude Code 能力的可复用、可调用的提示模板——技能是什么、如何编写、存放在哪里以及如何使用 /skill-name 调用它们。

Enterprise-focused AI platform specializing in embeddings, reranking, and RAG for search and information retrieval at scale.

Memory patterns for chat agents — buffer, summary, vector, and entity memory.

An overview of data pipelines in the ML context — batch vs streaming, ETL vs ELT, data quality, and schema validation.

Git for data and models — versioning datasets, pipelines, and experiments alongside source code.

How to systematically log, compare, and reproduce ML experiments using tracking tools.

用于模型、数据集和管道的平台与库。

Framework for LLM applications and agents.

Stateful agent graphs built on LangChain, where nodes are Python functions, edges define routing, and a shared TypedDict state enables cycles, conditional branching, persistence, and human-in-the-loop checkpoints.

用于大语言模型应用和 RAG 的数据框架。

Open-source platform for the complete ML lifecycle, covering experiment tracking, projects, models, and the registry.

Centralized store for versioning, staging, and governing ML model artifacts across their full lifecycle.

Multiple agents collaborating or competing.

跨平台、高性能的 ONNX 模型推理引擎，支持 CPU、GPU 和 NPU 执行提供程序。

Architecture where one LLM creates a step-by-step plan and another executes each step independently.

具有动态计算图的深度学习框架。

使用 TorchScript 和下一代 ExecuTorch 运行时在移动和边缘设备上部署 PyTorch 模型。

Components and design choices in RAG systems.

Example RAG pipelines and code snippets.

在智能体中交织推理和行动。

How LLMs and agents structure reasoning and action.

Combining retrieval with LLM generation for accurate, grounded answers.

Spec-driven reasoning pattern combining retrieval and decision design.

LLM 的实时 token 流式输出。

Google 的深度学习框架。

用于在 Android、iOS、嵌入式系统和微控制器上进行设备端机器学习推理的轻量级运行时。

分支推理以探索多条思维路径。

Storing and searching embeddings for RAG.

Cloud-native MLOps platform for experiment tracking, hyperparameter sweeps, artifact management, and collaborative reporting.

Claude Code 如何在长会话中管理上下文窗口——自动压缩、对话历史策略以及在大规模场景中保持会话有效性的实用技巧。

用于空间和图像数据的 CNN。

RNN 与序列数据。

将预训练模型适配到特定任务或领域。

Claude Code 中的扩展思考——它是什么、努力级别如何影响推理深度与速度，以及如何为不同任务类型配置思考行为。

连接 AI 模型与外部工具、数据源和服务的开放标准——实现任何 AI 应用程序中可移植、可互操作的工具使用。