Datasets:

cy0307
/

awesome-ai-native

category stringclasses 18 values	category_name stringclasses 18 values	group stringclasses 6 values	group_name stringclasses 6 values	title stringlengths 2 50	url stringlengths 15 77	author stringlengths 4 29	note stringlengths 36 128	note_zh stringlengths 11 52
foundations	Foundations & Concepts	concepts	Concepts & Theory	What is AI Native? (this repo)	https://github.com/ChaoYue0307/awesome-ai-native/blob/main/docs/concepts.md	This repo	Our own primer on the paradigm, the AI-Enabled vs. AI Native split, and the core shifts.	我们自己写的范式入门:AI-Enabled vs AI Native 的区别,以及几个核心转变。
foundations	Foundations & Concepts	concepts	Concepts & Theory	Software 2.0	https://karpathy.medium.com/software-2-0-a64152b37c35	Andrej Karpathy	The foundational argument that neural nets are a new kind of software where data and weights replace hand-written logic.	奠基性论述:神经网络是一种新软件,用数据和权重取代手写逻辑。
foundations	Foundations & Concepts	concepts	Concepts & Theory	Software Is Changing (Again) — Software 3.0	https://www.youtube.com/watch?v=LCEmiRjPEtQ	Andrej Karpathy	LLMs as a new programmable layer; the hottest new programming language is English.	把 LLM 看作新的可编程层;最热门的新编程语言是英语。
foundations	Foundations & Concepts	concepts	Concepts & Theory	[1hr Talk] Intro to Large Language Models	https://www.youtube.com/watch?v=zjkBMFhNj_g	Andrej Karpathy	The single best one-hour explainer of what an LLM is and how it behaves.	关于 LLM 是什么、如何表现的最佳一小时讲解。
foundations	Foundations & Concepts	concepts	Concepts & Theory	Emerging Architectures for LLM Applications	https://a16z.com/emerging-architectures-for-llm-applications/	a16z	A widely-cited reference architecture for the LLM app stack (orchestration, retrieval, tools).	被广泛引用的 LLM 应用栈参考架构(编排、检索、工具)。
foundations	Foundations & Concepts	concepts	Concepts & Theory	The Bitter Lesson	http://www.incompleteideas.net/IncIdeas/BitterLesson.html	Rich Sutton	Why general methods that leverage compute beat hand-crafted knowledge — the intellectual backbone of the AI-native bet.	为什么靠算力的通用方法终将胜过手工知识——AI 原生这一赌注的思想根基。
foundations	Foundations & Concepts	concepts	Concepts & Theory	Generative AI Exists Because of the Transformer	https://ig.ft.com/generative-ai/	Financial Times	A gorgeous visual explainer of how transformers actually work — the best non-technical intro.	极其精美的可视化讲解,说明 Transformer 究竟如何工作——最好的非技术入门。
patterns	Architecture & Patterns	concepts	Concepts & Theory	Architecture & Patterns (this repo)	https://github.com/ChaoYue0307/awesome-ai-native/blob/main/docs/patterns.md	This repo	Our catalog: model-as-runtime, probabilistic design, agent patterns, context patterns, reliability patterns.	我们的模式目录:模型即运行时、概率性设计、Agent 模式、上下文模式、可靠性模式。
patterns	Architecture & Patterns	concepts	Concepts & Theory	Building Effective Agents	https://www.anthropic.com/engineering/building-effective-agents	Anthropic	A clear taxonomy of agentic patterns (prompt chaining, routing, orchestrator-workers, evaluator-optimizer) and when to use each.	清晰的 Agent 模式分类法(提示链、路由、编排者-工人、评估者-优化者)及适用场景。
patterns	Architecture & Patterns	concepts	Concepts & Theory	LLM Powered Autonomous Agents	https://lilianweng.github.io/posts/2023-06-23-agent/	Lilian Weng	The reference deep-dive on agent architecture: planning, memory, and tool use.	关于 Agent 架构的参考级深度文章:规划、记忆与工具使用。
patterns	Architecture & Patterns	concepts	Concepts & Theory	Agentic Design Patterns	https://www.deeplearning.ai/the-batch/how-agents-can-improve-llm-performance/	Andrew Ng / DeepLearning.AI	Four foundational agentic patterns: reflection, tool use, planning, and multi-agent collaboration.	四种基础 Agent 模式:反思、工具使用、规划、多智能体协作。
patterns	Architecture & Patterns	concepts	Concepts & Theory	Patterns for Building LLM-based Systems & Products	https://eugeneyan.com/writing/llm-patterns/	Eugene Yan	A deeply practical field guide: evals, RAG, fine-tuning, caching, guardrails, and UX patterns.	极其务实的实战指南:评估、RAG、微调、缓存、护栏与 UX 模式。
agents	Agents	build	Building Blocks	Building Effective Agents	https://www.anthropic.com/engineering/building-effective-agents	Anthropic	The best practical starting point; argues for simple, composable patterns over heavy frameworks.	最佳实践起点,主张用简单可组合的模式而非笨重框架。
agents	Agents	build	Building Blocks	ReAct: Synergizing Reasoning and Acting	https://arxiv.org/abs/2210.03629	Yao et al.	The reason-act-observe loop underlying most agents.	多数 Agent 背后的“推理-行动-观察”循环。
agents	Agents	build	Building Blocks	Reflexion	https://arxiv.org/abs/2303.11366	Shinn et al.	Agents that improve by reflecting on past failures in natural language.	通过用自然语言反思过往失败来自我改进的 Agent。
agents	Agents	build	Building Blocks	Toolformer	https://arxiv.org/abs/2302.04761	Schick et al.	Models that teach themselves when and how to call APIs.	让模型自学何时、如何调用 API。
agents	Agents	build	Building Blocks	Generative Agents (Smallville)	https://arxiv.org/abs/2304.03442	Park et al.	25 agents with memory, planning, and reflection living in a simulated town — a landmark in agent believability.	25 个具备记忆、规划与反思的 Agent 生活在模拟小镇——Agent 可信度的里程碑。
agents	Agents	build	Building Blocks	12-Factor Agents	https://github.com/humanlayer/12-factor-agents	HumanLayer	Principles for production-grade LLM agents — the “12-factor app” idea applied to agents.	构建可上生产的 LLM Agent 的原则——把“12-factor app”思想用到 Agent 上。
agents	Agents	build	Building Blocks	Model Context Protocol (MCP)	https://modelcontextprotocol.io/	Anthropic	Open standard for connecting models to tools and data sources — the USB-C of AI integrations.	连接模型与工具/数据的开放标准——AI 集成的“USB-C”。
agents	Agents	build	Building Blocks	LangGraph	https://www.langchain.com/langgraph	LangChain	Low-level orchestration framework for stateful, multi-actor agent workflows as graphs.	用于有状态、多角色 Agent 工作流的底层编排框架(以图建模)。
agents	Agents	build	Building Blocks	OpenAI Agents SDK	https://openai.github.io/openai-agents-python/	OpenAI	Lightweight framework for multi-agent workflows with handoffs and guardrails.	用于多智能体工作流的轻量框架,支持交接与护栏。
agents	Agents	build	Building Blocks	AutoGen	https://microsoft.github.io/autogen/	Microsoft	Framework for multi-agent conversation and orchestration.	面向多智能体对话与编排的框架。
agents	Agents	build	Building Blocks	CrewAI	https://www.crewai.com/	CrewAI	Framework for orchestrating role-playing, collaborative agent crews.	编排“角色扮演”协作 Agent 团队的框架。
agents	Agents	build	Building Blocks	smolagents	https://github.com/huggingface/smolagents	Hugging Face	A minimal library for agents that think in code — tiny and readable.	构建“用代码思考”的 Agent 的极简库——小而易读。
agents	Agents	build	Building Blocks	Pydantic AI	https://ai.pydantic.dev/	Pydantic	Type-safe agent framework from the Pydantic team; strong structured outputs.	Pydantic 团队出品的类型安全 Agent 框架,结构化输出能力强。
agents	Agents	build	Building Blocks	OpenHands	https://github.com/All-Hands-AI/OpenHands	All Hands AI	Open-source platform for AI software-engineering agents (formerly OpenDevin).	面向 AI 软件工程 Agent 的开源平台(前身 OpenDevin)。
agents	Agents	build	Building Blocks	SWE-bench	https://www.swebench.com/	Princeton & Stanford	The benchmark that measures whether coding agents can resolve real GitHub issues.	衡量编码 Agent 能否解决真实 GitHub issue 的基准。
context	Context Engineering, RAG & Memory	build	Building Blocks	Retrieval-Augmented Generation (RAG)	https://arxiv.org/abs/2005.11401	Lewis et al.	The original paper pairing retrieval with generation.	把检索与生成结合的奠基论文。
context	Context Engineering, RAG & Memory	build	Building Blocks	Lost in the Middle	https://arxiv.org/abs/2307.03172	Liu et al.	Models ignore information buried in the middle of long contexts — essential reading for context design.	模型会忽略长上下文中间部分的信息——上下文设计的必读。
context	Context Engineering, RAG & Memory	build	Building Blocks	Retrieval-Augmented Generation: A Survey	https://arxiv.org/abs/2312.10997	Gao et al.	A comprehensive map of the RAG landscape, from naive to advanced and modular.	RAG 全景综述,从朴素到进阶再到模块化。
context	Context Engineering, RAG & Memory	build	Building Blocks	Introducing Contextual Retrieval	https://www.anthropic.com/news/contextual-retrieval	Anthropic	A simple technique that dramatically cuts retrieval failures — practical and benchmarked.	一个能大幅降低检索失败率的简单技术——务实且有基准。
context	Context Engineering, RAG & Memory	build	Building Blocks	MemGPT / Letta	https://www.letta.com/	Letta	Treating the context window like an OS manages memory — agents with long-term recall.	把上下文窗口当作操作系统管理内存——具备长期记忆的 Agent。
context	Context Engineering, RAG & Memory	build	Building Blocks	LlamaIndex	https://www.llamaindex.ai/	LlamaIndex	Data framework for connecting LLMs to your data; strong RAG primitives.	连接 LLM 与你的数据的框架,RAG 原语强大。
context	Context Engineering, RAG & Memory	build	Building Blocks	Chroma	https://www.trychroma.com/	Chroma	Open-source embedding database for building retrieval into apps.	开源向量数据库,把检索能力嵌入应用。
context	Context Engineering, RAG & Memory	build	Building Blocks	Weaviate	https://weaviate.io/	Weaviate	Open-source vector database with hybrid search and built-in modules.	支持混合检索与内置模块的开源向量数据库。
context	Context Engineering, RAG & Memory	build	Building Blocks	Qdrant	https://qdrant.tech/	Qdrant	High-performance vector database written in Rust.	用 Rust 编写的高性能向量数据库。
context	Context Engineering, RAG & Memory	build	Building Blocks	Pinecone Learn	https://www.pinecone.io/learn/	Pinecone	High-quality tutorials on embeddings, vector search, and RAG.	关于嵌入、向量检索与 RAG 的高质量教程。
prompting	Prompting & Reasoning	build	Building Blocks	Prompt Engineering Guide	https://www.promptingguide.ai/	DAIR.AI	Comprehensive, well-organized reference for prompting techniques.	全面、条理清晰的提示技巧参考。
prompting	Prompting & Reasoning	build	Building Blocks	Anthropic Prompt Engineering	https://docs.claude.com/en/docs/build-with-claude/prompt-engineering/overview	Anthropic	Official, practical guidance for prompting Claude well.	官方的、务实的 Claude 提示指南。
prompting	Prompting & Reasoning	build	Building Blocks	OpenAI Prompt Engineering Guide	https://platform.openai.com/docs/guides/prompt-engineering	OpenAI	Official tactics for prompting GPT models effectively.	高效提示 GPT 模型的官方策略。
prompting	Prompting & Reasoning	build	Building Blocks	Prompt Engineering (post)	https://lilianweng.github.io/posts/2023-03-15-prompt-engineering/	Lilian Weng	A rigorous survey-style walkthrough of prompting methods and why they work.	严谨的综述式梳理:各种提示方法及其奏效原因。
prompting	Prompting & Reasoning	build	Building Blocks	Chain-of-Thought Prompting	https://arxiv.org/abs/2201.11903	Wei et al.	Eliciting step-by-step reasoning dramatically improves complex tasks.	引导逐步推理可显著提升复杂任务表现。
prompting	Prompting & Reasoning	build	Building Blocks	Self-Consistency	https://arxiv.org/abs/2203.11171	Wang et al.	Sample multiple reasoning paths and take a majority vote.	采样多条推理路径再多数表决。
prompting	Prompting & Reasoning	build	Building Blocks	Tree of Thoughts	https://arxiv.org/abs/2305.10601	Yao et al.	Search over a tree of reasoning steps with lookahead and backtracking.	在推理步骤树上做带前瞻与回溯的搜索。
prompting	Prompting & Reasoning	build	Building Blocks	The Prompt Report	https://arxiv.org/abs/2406.06608	Schulhoff et al.	A systematic survey cataloguing 58+ prompting techniques.	系统性综述,梳理 58 余种提示技巧。
ux	AI UX & Interface Patterns	build	Building Blocks	Shape of AI	https://www.shapeof.ai/	Emily Campbell	A pattern library of UX patterns for AI products — wayfinders, trust, governors, and more.	面向 AI 产品的 UX 模式库——引导、信任、约束等模式。
ux	AI UX & Interface Patterns	build	Building Blocks	NN/g: AI	https://www.nngroup.com/topic/ai/	Nielsen Norman Group	Research-grounded UX guidance for generative AI and chat interfaces.	基于研究的生成式 AI 与对话界面 UX 指南。
ux	AI UX & Interface Patterns	build	Building Blocks	Generative UI (Vercel AI SDK)	https://sdk.vercel.ai/docs/ai-sdk-rsc/generative-ui-state	Vercel	How to stream model-generated React components as the interface itself.	如何把模型生成的 React 组件作为界面本身进行流式渲染。
eval	Evaluation & Reliability	quality	Quality & Operations	Your AI Product Needs Evals	https://hamel.dev/blog/posts/evals/	Hamel Husain	The canonical case for evals and how to actually build them — start here.	关于为什么以及怎么真正做 evals 的权威文章——从这里开始。
eval	Evaluation & Reliability	quality	Quality & Operations	OpenAI Evals	https://github.com/openai/evals	OpenAI	A framework and open registry for evaluating LLMs and systems.	评估 LLM 与系统的框架及开放注册表。
eval	Evaluation & Reliability	quality	Quality & Operations	Ragas	https://github.com/explodinggradients/ragas	Exploding Gradients	Metrics and tooling specifically for evaluating RAG pipelines.	专门评估 RAG 流水线的指标与工具。
eval	Evaluation & Reliability	quality	Quality & Operations	promptfoo	https://www.promptfoo.dev/	promptfoo	Open-source tool for testing and red-teaming prompts and LLM apps.	测试与红队评估 prompt 和 LLM 应用的开源工具。
eval	Evaluation & Reliability	quality	Quality & Operations	DeepEval	https://github.com/confident-ai/deepeval	Confident AI	A pytest-like framework for unit-testing LLM outputs.	类 pytest 的框架,用来对 LLM 输出做单元测试。
eval	Evaluation & Reliability	quality	Quality & Operations	Inspect	https://inspect.aisi.org.uk/	UK AI Safety Institute	A serious, open-source framework for evaluating LLMs, built for rigor.	严谨、开源的 LLM 评估框架,为可靠性而生。
eval	Evaluation & Reliability	quality	Quality & Operations	HELM	https://crfm.stanford.edu/helm/	Stanford CRFM	A holistic, transparent benchmark across many models and scenarios.	覆盖众多模型与场景的整体、透明基准。
eval	Evaluation & Reliability	quality	Quality & Operations	LMSYS Chatbot Arena	https://lmarena.ai/	LMSYS	Crowdsourced, head-to-head model rankings via blind human preference votes.	通过盲测人类偏好投票得到的众包模型对战排行。
ops	Observability & LLMOps	quality	Quality & Operations	LangSmith	https://www.langchain.com/langsmith	LangChain	Tracing, evaluation, and monitoring for LLM apps — framework-agnostic.	面向 LLM 应用的追踪、评估与监控,框架无关。
ops	Observability & LLMOps	quality	Quality & Operations	Langfuse	https://langfuse.com/	Langfuse	Open-source LLM engineering platform: tracing, evals, prompt management.	开源 LLM 工程平台:追踪、评估、提示管理。
ops	Observability & LLMOps	quality	Quality & Operations	Arize Phoenix	https://github.com/Arize-ai/phoenix	Arize AI	Open-source observability for LLM/agent apps with OpenTelemetry tracing.	面向 LLM/Agent 应用的开源可观测性,支持 OpenTelemetry 追踪。
ops	Observability & LLMOps	quality	Quality & Operations	Helicone	https://www.helicone.ai/	Helicone	Drop-in observability and cost monitoring via a proxy.	通过代理即插即用的可观测性与成本监控。
safety	Safety, Security & Governance	quality	Quality & Operations	OWASP Top 10 for LLM Applications	https://owasp.org/www-project-top-10-for-large-language-model-applications/	OWASP	The canonical security checklist for LLM apps — prompt injection, data leakage, and more.	LLM 应用的权威安全清单——提示注入、数据泄露等。
safety	Safety, Security & Governance	quality	Quality & Operations	Prompt Injection (series)	https://simonwillison.net/series/prompt-injection/	Simon Willison	The definitive running coverage of prompt injection — the field's hardest unsolved security problem.	关于提示注入最权威的持续报道——该领域最棘手的未解安全难题。
safety	Safety, Security & Governance	quality	Quality & Operations	NIST AI Risk Management Framework	https://www.nist.gov/itl/ai-risk-management-framework	NIST	A widely-referenced framework for managing AI risk across the lifecycle.	被广泛引用的全生命周期 AI 风险管理框架。
safety	Safety, Security & Governance	quality	Quality & Operations	AI Incident Database	https://incidentdatabase.ai/	Responsible AI Collaborative	A searchable record of real-world AI harms — learn from what has gone wrong.	可检索的真实 AI 事故记录——从已发生的失败中学习。
safety	Safety, Security & Governance	quality	Quality & Operations	Core Views on AI Safety	https://www.anthropic.com/news/core-views-on-ai-safety	Anthropic	A clear articulation of why and how to take frontier-model safety seriously.	清晰阐述为何以及如何认真对待前沿模型安全。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	LangChain	https://www.langchain.com/	LangChain	The most widely-used framework for composing LLM applications.	使用最广的 LLM 应用组合框架。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	LlamaIndex	https://www.llamaindex.ai/	LlamaIndex	Data/RAG framework for grounding LLMs in your data.	把 LLM 接到你数据上的数据/RAG 框架。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	Vercel AI SDK	https://sdk.vercel.ai/	Vercel	TypeScript toolkit for building streaming AI UIs and agents.	构建流式 AI 界面与 Agent 的 TypeScript 工具包。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	Semantic Kernel	https://github.com/microsoft/semantic-kernel	Microsoft	SDK for integrating LLMs into apps across C#, Python, and Java.	把 LLM 集成进应用的 SDK,支持 C#、Python、Java。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	DSPy	https://github.com/stanfordnlp/dspy	Stanford NLP	Program — don't prompt. Optimize LLM pipelines as compiled programs.	用编程而非手写 prompt:把 LLM 流水线当作可编译程序来优化。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	Instructor	https://python.useinstructor.com/	Jason Liu	Structured outputs from LLMs via Pydantic — reliable JSON, every time.	用 Pydantic 从 LLM 拿结构化输出——稳定的 JSON。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	Outlines	https://github.com/dottxt-ai/outlines	dottxt	Guaranteed-structured generation via constrained decoding.	通过受限解码实现保证结构化的生成。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	LiteLLM	https://github.com/BerriAI/litellm	BerriAI	One OpenAI-compatible interface to 100+ model providers.	用一套 OpenAI 兼容接口调用 100+ 模型厂商。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	Ollama	https://ollama.com/	Ollama	Run open-weight LLMs locally with a single command.	一条命令在本地运行开源权重 LLM。
tools	Tools, Frameworks & SDKs	stack	Models & Tooling	Pydantic	https://docs.pydantic.dev/	Pydantic	The data-validation backbone behind reliable structured LLM I/O in Python.	支撑 Python 中可靠结构化 LLM 输入输出的数据校验基石。
models	Models & APIs	stack	Models & Tooling	Anthropic API (Claude)	https://docs.claude.com/	Anthropic	Docs for the Claude model family; strong at coding, agents, and long context.	Claude 模型族文档;擅长编码、Agent 与长上下文。
models	Models & APIs	stack	Models & Tooling	OpenAI API	https://platform.openai.com/docs	OpenAI	Docs for the GPT model family and tooling.	GPT 模型族与工具文档。
models	Models & APIs	stack	Models & Tooling	Google Gemini API	https://ai.google.dev/	Google	Docs for the Gemini multimodal model family.	Gemini 多模态模型族文档。
models	Models & APIs	stack	Models & Tooling	DeepSeek	https://www.deepseek.com/	DeepSeek	Open, strong, and cheap reasoning models that reshaped cost expectations.	开放、强大且低价的推理模型,重塑了成本预期。
models	Models & APIs	stack	Models & Tooling	Cohere	https://cohere.com/	Cohere	Enterprise-focused models with strong embeddings and reranking.	面向企业的模型,嵌入与重排能力强。
models	Models & APIs	stack	Models & Tooling	Hugging Face	https://huggingface.co/	Hugging Face	The hub for open models, datasets, and the Transformers library.	开源模型、数据集与 Transformers 库的中心。
models	Models & APIs	stack	Models & Tooling	OpenRouter	https://openrouter.ai/	OpenRouter	One API and marketplace across many model providers.	横跨众多厂商的统一 API 与市场。
models	Models & APIs	stack	Models & Tooling	Replicate	https://replicate.com/	Replicate	Run and fine-tune open models via a simple API.	通过简单 API 运行并微调开源模型。
models	Models & APIs	stack	Models & Tooling	Groq	https://groq.com/	Groq	Ultra-low-latency inference for open models.	面向开源模型的超低延迟推理。
open-models	Open Models & Fine-tuning	stack	Models & Tooling	Llama	https://www.llama.com/	Meta	The open-weight model family that catalyzed the local/open ecosystem.	催化本地/开源生态的开源权重模型族。
open-models	Open Models & Fine-tuning	stack	Models & Tooling	Mistral AI	https://mistral.ai/	Mistral	Efficient open-weight European models with strong performance-per-parameter.	高效的欧洲开源权重模型,单位参数性能强。
open-models	Open Models & Fine-tuning	stack	Models & Tooling	Unsloth	https://unsloth.ai/	Unsloth	Fine-tune open LLMs 2x faster with far less memory.	以 2 倍速度、远更低显存微调开源 LLM。
open-models	Open Models & Fine-tuning	stack	Models & Tooling	TRL	https://github.com/huggingface/trl	Hugging Face	Train and align transformers with SFT, DPO, and RLHF.	用 SFT、DPO、RLHF 训练并对齐 transformer。
open-models	Open Models & Fine-tuning	stack	Models & Tooling	vLLM	https://github.com/vllm-project/vllm	vLLM	High-throughput, memory-efficient inference and serving engine.	高吞吐、省显存的推理与服务引擎。
open-models	Open Models & Fine-tuning	stack	Models & Tooling	QLoRA	https://arxiv.org/abs/2305.14314	Dettmers et al.	Fine-tune large models on a single GPU via 4-bit quantization — a democratizing result.	通过 4-bit 量化在单张 GPU 上微调大模型——具普惠意义的成果。
multimodal	Multimodal & Creative AI	stack	Models & Tooling	Midjourney	https://www.midjourney.com/	Midjourney	Best-in-class text-to-image generation with a distinctive aesthetic.	一流的文生图,风格独特。
multimodal	Multimodal & Creative AI	stack	Models & Tooling	ElevenLabs	https://elevenlabs.io/	ElevenLabs	State-of-the-art voice synthesis and cloning.	顶尖的语音合成与声音克隆。
multimodal	Multimodal & Creative AI	stack	Models & Tooling	Suno	https://suno.com/	Suno	Generate full songs — vocals and instruments — from a prompt.	从一句话生成完整歌曲——人声与乐器俱全。
multimodal	Multimodal & Creative AI	stack	Models & Tooling	Runway	https://runwayml.com/	Runway	Generative video tools used in real film and creative production.	用于真实影视与创意制作的生成式视频工具。
dev	AI Native Development	practice	Practice & Products	Claude Code Best Practices	https://www.anthropic.com/engineering/claude-code-best-practices	Anthropic	How to actually work effectively with an agentic coding tool.	如何高效使用 Agent 编码工具的实战经验。
dev	AI Native Development	practice	Practice & Products	Simon Willison's Weblog	https://simonwillison.net/	Simon Willison	The best running commentary on practical LLM application development.	关于 LLM 应用实战开发的最佳持续评论。
dev	AI Native Development	practice	Practice & Products	The Rise of the AI Engineer	https://www.latent.space/p/ai-engineer	swyx / Latent Space	The essay that named the AI Engineer role and its emerging discipline.	为“AI 工程师”这一角色命名并定义其新兴学科的文章。
dev	AI Native Development	practice	Practice & Products	What We Learned Building with LLMs (Part 1)	https://applied-llms.org/	Yan, Bornstein, Husain et al.	Hard-won tactical, operational, and strategic lessons from shipping LLM products.	把 LLM 产品做出来过程中得来的战术、运营与战略经验。
dev	AI Native Development	practice	Practice & Products	Aider	https://aider.chat/	Paul Gauthier	AI pair programming in your terminal, with deep git integration.	终端里的 AI 结对编程,深度集成 git。
dev	AI Native Development	practice	Practice & Products	Continue	https://www.continue.dev/	Continue	Open-source AI code assistant you can fully customize.	可完全自定义的开源 AI 编码助手。

End of preview. Expand in Data Studio

Awesome AI Native — Dataset

A curated, structured dataset of 137 resources across 18 categories and 6 themes for understanding and building AI Native products, systems, and teams — where large models are the foundation, not a feature.

This is the machine-readable companion to the Awesome AI Native list and its website.

Columns

Field	Description
`category`	Category id (e.g. `agents`, `eval`)
`category_name`	Human-readable category name
`group`	Theme id grouping related categories
`group_name`	Human-readable theme (e.g. `Building Blocks`)
`title`	Resource title
`url`	Canonical link to the resource
`author`	Author or organization
`note`	One-line English description of why it matters
`note_zh`	One-line Simplified Chinese description (may be empty)

Usage

from datasets import load_dataset

ds = load_dataset("cy0307/awesome-ai-native", split="train")
print(ds[0])

Themes

Concepts & Theory — Foundations & Concepts, Architecture & Patterns
Building Blocks — Agents, Context Engineering, RAG & Memory, Prompting & Reasoning, AI UX & Interface Patterns
Quality & Operations — Evaluation & Reliability, Observability & LLMOps, Safety, Security & Governance
Models & Tooling — Tools, Frameworks & SDKs, Models & APIs, Open Models & Fine-tuning, Multimodal & Creative AI
Practice & Products — AI Native Development, AI Native Products
Reference & Community — Papers & Deep Reads, Courses & Learning, Talks, Podcasts & Newsletters

Source & maintenance

Generated from data/resources.json in the source repo via node scripts/build.mjs. To suggest a resource, open a PR there.

License

CC0 1.0 — public domain. Links point to third-party resources owned by their respective authors.

Downloads last month: -