AI Agent learning center
Official tutorials, courses, engineering guides, and frontier papers—checked daily and sorted by original publication date.
Latest automated source release Nov 14, 2026
Latest additions
Daily scans of official tutorials, courses, GitHub, and arXiv, with the newest releases first.
WeightBench: Evaluating Language Models as Weight-Space Programmers
A verified paper resource from arxiv.
arxiv · paperImitation Bootstrapped Multi-Agent Reinforcement Learning
A verified paper resource from arxiv.
arxiv · paperModHarness-RL: Recasting Harnesses as Learnable and Evolving Programs for Agentic RL
A verified paper resource from arxiv.
arxiv · paperAlignment by Socialization: Learning Tradeoffs Under Selection
A verified paper resource from arxiv.
arxiv · paperCan Mechanistic Interpretability Help Explain LLM Agent Failures? A Case Study on Web Element Hallucination
A verified paper resource from arxiv.
arxiv · paperAgentic AI for Discovering Fault-Tolerant Quantum Circuits
A verified paper resource from arxiv.
arxiv · paperTradeGauntlet: Evaluating Compliance Controls for Autonomous Trading Agents
A verified paper resource from arxiv.
arxiv · paperInteractive Program Induction with Language Model Priors
A verified paper resource from arxiv.
arxiv · paperFrom AI-Enabled Remote Patient Monitoring Toward Agentic Robotics: Embodiment, Governance, and Care Integration
A verified paper resource from arxiv.
Leading-company whitepapers and engineering guides
Free, authoritative first-party methods from Anthropic, OpenAI, and Google.
Building Effective Agents
Official guidance from Anthropic.
AnthropicEffective Context Engineering for AI Agents
Official guidance from Anthropic.
AnthropicBuilding Effective AI Agents: Architecture Patterns(PDF)
Official guidance from Anthropic.
OpenAIA Practical Guide to Building Agents(PDF)
Official guidance from OpenAI.
GoogleAgents Whitepaper(Kaggle 系列)
Official guidance from Google.
Anthropic(开源)Model Context Protocol(MCP)官方文档
Official guidance from Anthropic(开源).
Google(开源)A2A(Agent2Agent)协议
Official guidance from Google(开源).
Stanford HAIAI Index Report
Official guidance from Stanford HAI.
Nathan Benaich / Air StreetState of AI Report
Official guidance from Nathan Benaich / Air Street.
Frontier courses from leading universities
Free public courses from Berkeley, Stanford, MIT, CMU, and other leading universities.
CS294/194-196 Agentic AI(Fall 2025,免费 MOOC)
A public course from UC Berkeley.
UC Berkeley RDI · 美国 · 视频公开LLM Agents MOOC(Berkeley 公开版入口)
A public course from UC Berkeley RDI.
UC Berkeley · 美国 · 视频公开CS294/194-280 Advanced LLM Agents(Spring 2025)
A public course from UC Berkeley.
UC Berkeley · 美国 · 视频公开CS294/194-196 LLM Agents(Fall 2024)
A public course from UC Berkeley.
Stanford · 美国CS329A Self-Improving AI Agents
A public course from Stanford.
Stanford · 美国 · 视频公开CS25 Transformers United(YouTube 全公开)
A public course from Stanford.
MIT · 美国AI Agent Index
A public course from MIT.
CMU · 美国 · 视频公开Advanced NLP(Graham Neubig,YouTube 全公开)
A public course from CMU.
台湾大学 · 中文 · 视频公开李宏毅《生成式 AI 导论》(含 Agent 专题)
A public course from 台湾大学.
李沐 / 亚马逊 · 中文 · 视频公开《动手学深度学习》d2l.ai
A public course from 李沐 / 亚马逊.
Datawhale 社区 · 中文Datawhale 开源课程(wow-agent / self-llm)
A public course from Datawhale 社区.
清华大学 NLP 实验室 · 中文THUNLP / OpenBMB 开源体系(ChatDev · AgentVerse · XAgent)
A public course from 清华大学 NLP 实验室.
Free courses and video tutorials
Start at no cost with material from Hugging Face, DeepLearning.AI, Microsoft, Anthropic, and Google.
AI Agents Course(免费 + 证书)
A learning resource from Hugging Face.
DeepLearning.AI(吴恩达)Agentic AI 及 Agent 系列短课
A learning resource from DeepLearning.AI(吴恩达).
MicrosoftAI Agents for Beginners(含中文翻译)
A learning resource from Microsoft.
Anthropic · 可拿证书Anthropic Academy / Claude 官方课程
A learning resource from Anthropic.
OpenAIOpenAI Academy
A learning resource from OpenAI.
LangChain · 可拿证书LangChain Academy(LangGraph 免费课)
A learning resource from LangChain.
Google × Kaggle · 可拿证书5-Day GenAI Intensive(含 Agents 白皮书日)
A learning resource from Google × Kaggle.
Analytics Vidhya(聚合)15 门免费 AI Agent 认证课程合集(2026 版)
A learning resource from Analytics Vidhya(聚合).
Research papers: frontier work and essential classics
Track current surveys while covering the papers practitioners should know.
Frontier work (2025–2026 surveys)
Rethinking Agentic Reinforcement Learning in LLMs
Open the original paper for the abstract and full research details.
2026 · arXiv:2603.22862The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestration
Open the original paper for the abstract and full research details.
2026 · arXiv:2603.22386From Static Templates to Dynamic Runtime Graphs: Workflow Optimization for LLM Agents
Open the original paper for the abstract and full research details.
2025 · arXiv:2508.12752Deep Research: A Survey of Autonomous Research Agents
Open the original paper for the abstract and full research details.
2025 · arXiv:2509.06733Reinforcement Learning Foundations for Deep Research Systems: A Survey
Open the original paper for the abstract and full research details.
2025 · arXiv:2510.16724A Comprehensive Survey on RL-based Agentic Search
Open the original paper for the abstract and full research details.
2025 · arXiv:2503.12434A Survey on the Optimization of LLM-based Agents
Open the original paper for the abstract and full research details.
2025 · arXiv:2502.14321Beyond Self-Talk: A Communication-Centric Survey of LLM-Based Multi-Agent Systems
Open the original paper for the abstract and full research details.
2025 · arXiv:2510.04023LLM-Based Data Science Agents: A Survey
Open the original paper for the abstract and full research details.
Essential classics
ReAct: Synergizing Reasoning and Acting in Language Models
Open the original paper for the abstract and full research details.
2023 · arXiv:2302.04761Toolformer: Language Models Can Teach Themselves to Use Tools
Open the original paper for the abstract and full research details.
2023 · arXiv:2303.11366Reflexion: Language Agents with Verbal Reinforcement Learning
Open the original paper for the abstract and full research details.
2023 · arXiv:2304.03442Generative Agents: Interactive Simulacra of Human Behavior
Open the original paper for the abstract and full research details.
2023 · arXiv:2305.10601Tree of Thoughts: Deliberate Problem Solving with LLMs
Open the original paper for the abstract and full research details.
2023 · arXiv:2305.16291Voyager: An Open-Ended Embodied Agent with LLMs
Open the original paper for the abstract and full research details.
2023 · arXiv:2308.08155AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation
Open the original paper for the abstract and full research details.
2023 · arXiv:2308.00352MetaGPT: Meta Programming for Multi-Agent Collaborative Framework
Open the original paper for the abstract and full research details.
2023 · arXiv:2308.11432A Survey on LLM-based Autonomous Agents
Open the original paper for the abstract and full research details.
2023 · arXiv:2309.07864The Rise and Potential of LLM-Based Agents: A Survey
Open the original paper for the abstract and full research details.
2023 · arXiv:2310.06770SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
Open the original paper for the abstract and full research details.
2023 · arXiv:2307.13854WebArena: A Realistic Web Environment for Building Autonomous Agents
Open the original paper for the abstract and full research details.
2023 · arXiv:2311.12983GAIA: A Benchmark for General AI Assistants
Open the original paper for the abstract and full research details.
2024 · arXiv:2406.12045τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
Open the original paper for the abstract and full research details.
Essential skill tree
Ten knowledge nodes that form a practical growth path for agent engineers.
Learning module 01
Open the linked resources to build this part of your agent-engineering practice.
Learning module 02
Open the linked resources to build this part of your agent-engineering practice.
Learning module 03
Open the linked resources to build this part of your agent-engineering practice.
Learning module 04
Open the linked resources to build this part of your agent-engineering practice.
Learning module 05
Open the linked resources to build this part of your agent-engineering practice.
Learning module 06
Open the linked resources to build this part of your agent-engineering practice.
Learning module 07
Open the linked resources to build this part of your agent-engineering practice.
Learning module 08
Open the linked resources to build this part of your agent-engineering practice.
Learning module 09
Open the linked resources to build this part of your agent-engineering practice.
Learning module 10
Open the linked resources to build this part of your agent-engineering practice.
Daily updates to agent learning resources
New tutorials, courses, and papers are collected by publication date.
Subscribe for updates →