awesome-agent-learning

Guides, courses & reading lists for learning to build autonomous LLM agents
https://github.com/artnitolog/awesome-agent-learning

Last synced: about 8 hours ago
JSON representation

Framework Tutorials
- Hugging Face smolagents Tutorials
- LangChain: Build an Agent - on guide by Microsoft that demonstrates how to build an AI assistant capable of tool use with minimal setup.
- LlamaIndex Learning
- Microsoft AutoGen AgentChat Quickstart & Tutorial
- OpenAI Agents SDK Examples - world applications, providing examples for common design patterns, tool usage, and multi-turn orchestration.
- OpenAI Assistants API - by-step tutorial on building AI assistants using OpenAI's Assistants API. It covers creating assistants with specific instructions, integrating tools like code execution and file search, and managing conversations.
- CrewAI Quickstart Guide - on tutorial walking through building a multi-agent system using CrewAI framework. It demonstrates how to define different agent roles, assign tasks and orchestrate their collaboration.
- Haystack Tutorials
- AutoGPT Guide - by-step tutorial for setting up the full AutoGPT platform locally. Covering both backend and frontend configurations, enabling developers to build and manage AI agents.
- AutoGPT Guide - by-step tutorial for setting up the full AutoGPT platform locally. Covering both backend and frontend configurations, enabling developers to build and manage AI agents.
- CrewAI Quickstart Guide - on tutorial walking through building a multi-agent system using CrewAI framework. It demonstrates how to define different agent roles, assign tasks and orchestrate their collaboration.
- Haystack Tutorials
- Hugging Face smolagents Tutorials
- LangChain: Build an Agent - on guide by Microsoft that demonstrates how to build an AI assistant capable of tool use with minimal setup.
- OpenAI Agents SDK Examples - world applications, providing examples for common design patterns, tool usage, and multi-turn orchestration.
- OpenAI Assistants API - by-step tutorial on building AI assistants using OpenAI's Assistants API. It covers creating assistants with specific instructions, integrating tools like code execution and file search, and managing conversations.
- LlamaIndex Learning
- Microsoft AutoGen AgentChat Quickstart & Tutorial
Evaluation Benchmarks
- AgentBench - domain benchmark for measuring LLMs in autonomous agent roles, spanning 8 tasks (including OS management, SQL ops, web browsing/shopping, games, puzzles). Introduced a leaderboard for reproducible comparison.
- BrowseComp - crafted questions designed to assess agents ability to find hard-to-locate information across the web.
- GAIA - question benchmark that tests an agent's reasoning, web search, tool-use and multimodal understanding through short-answer tasks split into 3 difficulty levels.
- OSWorld - and desktop-tasks, including UI/GUI manipulations, and graded only by execution traces.
- SWE-bench - scale benchmark of 2k+ real-world github issues from Python repositories, where LLM agents must generate patches and pass tests, all inside a docker environment fully reproducible evaluation. Includes multiple extra versions: Lite with easy tasks, Verified with hand-validated issues, and Multimodal.
- ToolBench - task benchmark that asks agents to call real APIs from different web services, including weather, spreadsheets, shopping, reservations, and virtual environments.
- WebArena - hosted benchmark environment comprising 4 web applications: e‑commerce, forums, CMS, code — 800+ long-horizon tasks.
- AgentBench - domain benchmark for measuring LLMs in autonomous agent roles, spanning 8 tasks (including OS management, SQL ops, web browsing/shopping, games, puzzles). Introduced a leaderboard for reproducible comparison.
- BrowseComp - crafted questions designed to assess agents ability to find hard-to-locate information across the web.
- GAIA - question benchmark that tests an agent's reasoning, web search, tool-use and multimodal understanding through short-answer tasks split into 3 difficulty levels.
- OSWorld - and desktop-tasks, including UI/GUI manipulations, and graded only by execution traces.
- SWE-bench - scale benchmark of 2k+ real-world github issues from Python repositories, where LLM agents must generate patches and pass tests, all inside a docker environment fully reproducible evaluation. Includes multiple extra versions: Lite with easy tasks, Verified with hand-validated issues, and Multimodal.
- ToolBench - task benchmark that asks agents to call real APIs from different web services, including weather, spreadsheets, shopping, reservations, and virtual environments.
- WebArena - hosted benchmark environment comprising 4 web applications: e‑commerce, forums, CMS, code — 800+ long-horizon tasks.
Foundational Courses
- Agentic AI and AI Agents: A Primer for Leaders - technical executives and product managers who want to learn fundamentals of AI agents. Provides relevant theory and teaches no-code approaches of implementing AI agents using custom GPTs.
- Learn AI Agents Handbook
- Microsoft's AI Agents for Beginners - friendly but comprehensive open-source course comprising 11 lessons on building AI agents. Covers tool integration, RAG, agentic design patterns, multi-agent systems, deploying in production. Lessons include written materials, code samples and videos. Focused on Microsoft frameworks (Azure AI Agent Service,
- Multi AI Agent Systems with crewAI - friendly course teaching how to build and deploy AI agents using CrewAI framework. Covers basic concepts including tools management, memory organization, errors handling, agent cooperation. Introduces a lot of AI agent examples for common business processes.
- Advanced Large Language Model Agents - level course exploring the design and deployment of LLM-powered agents. Covers LLM foundations and infrastructure, reasoning, tool use, multi-agent collaboration, and various applications. Features guest lectures from leading researchers.
- Advanced Large Language Model Agents - level course exploring the design and deployment of LLM-powered agents. Covers LLM foundations and infrastructure, reasoning, tool use, multi-agent collaboration, and various applications. Features guest lectures from leading researchers.
- Agentic AI and AI Agents: A Primer for Leaders - technical executives and product managers who want to learn fundamentals of AI agents. Provides relevant theory and teaches no-code approaches of implementing AI agents using custom GPTs.
- Learn AI Agents Handbook
- Microsoft's AI Agents for Beginners - friendly but comprehensive open-source course comprising 11 lessons on building AI agents. Covers tool integration, RAG, agentic design patterns, multi-agent systems, deploying in production. Lessons include written materials, code samples and videos. Focused on Microsoft frameworks (Azure AI Agent Service,
- Multi AI Agent Systems with crewAI - friendly course teaching how to build and deploy AI agents using CrewAI framework. Covers basic concepts including tools management, memory organization, errors handling, agent cooperation. Introduces a lot of AI agent examples for common business processes.
- AI Agents Masterclass - friendly episodic series with full code walkthroughs to build AI agents. The course covers LangChain, LangGraph, RAG techniques, n8n workflow agents. Each episode is accompanied by the exact code used in the videos.
Conceptual Guides
- A Practical Guide to Building Agents (OpenAI) - by-step playbook offering best practices for designing autonomous AI agents. It covers identifying suitable problems, agent architecture, and measuring success.
- A Survey on Large Language Model based Autonomous Agents - based autonomous agents. The work also discusses evaluation strategies and outlines future research directions.
- Building effective agents (Anthropic) - depth guide by Anthropic offering practical strategies for designing LLM-based agents. Emphasizes starting with simple, composable patterns—like prompt chaining and routing—before progressing to more autonomous, tool-using agents.
- LLM Powered Autonomous Agents (Lilian Weng)
- What Are AI Agents? - by-step guide to building one. The article offers a practical walkthrough of building such agents with the focus on business architecture, the implementation is left to the readers.
- Building effective agents (Anthropic) - depth guide by Anthropic offering practical strategies for designing LLM-based agents. Emphasizes starting with simple, composable patterns—like prompt chaining and routing—before progressing to more autonomous, tool-using agents.
- LLM Powered Autonomous Agents (Lilian Weng)
- What Are AI Agents? - by-step guide to building one. The article offers a practical walkthrough of building such agents with the focus on business architecture, the implementation is left to the readers.
- A Practical Guide to Building Agents (OpenAI) - by-step playbook offering best practices for designing autonomous AI agents. It covers identifying suitable problems, agent architecture, and measuring success.
- A Survey on Large Language Model based Autonomous Agents - based autonomous agents. The work also discusses evaluation strategies and outlines future research directions.
Related Resources

Programming Languages

Python 9 Jupyter Notebook 2

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Awesome

awesome-agent-learning

Framework Tutorials

Evaluation Benchmarks

Foundational Courses

Conceptual Guides

awesome-agent-learning

Framework Tutorials

Evaluation Benchmarks

Foundational Courses

Conceptual Guides

Related Resources