Skip to content
English

Switch theme

Discover▾

Not sure where to start? This week’s hottest software, models and stacks in three minutes.

AI Apps▾

AI software people use: from assistants like ChatGPT to tools for coding, video and workflows, picked by purpose.

For Agents▾

Software for AI: tools that let your agent (Claude Code, Codex, Cursor…) browse, read docs, run code and remember.

Models▾

Choose models, estimate cost, self-host: prices, rankings, benchmarks, free tiers and compute.

Solutions▾

A goal or an error message? Ready-made stacks, steps and fixes by scenario.

Learn▾

A systematic course from zero to the frontier, with an algorithm library, must-read papers and tutorials.

Frontier▾

Daily AI news, weekly highlights, the latest papers and how AI is changing every discipline.

// Stage 8 · RAG, Agents & AI Engineering

Agents: Tool Use, MCP, Planning & Memory

ReAct, function calling, MCP, reflection, memory, multi-agent systems, computer use and sandboxes.

Read the lesson → Code & exercises ↓

about 12 hours10 algorithms4 paperslesson text ready

01Goals

Do these first: RAG Engineering & Evaluation · Prompting & Using AI Well

02Read on this site

03Algorithms to master

Each card has the original paper, prerequisites and resources.

ReAct (Reason + Act) Interleaves reasoning, actions and observations. 2022core Toolformer A model teaches itself when and how to call APIs. 2023ext Function Calling / Tool Use Models emit structured function calls that follow a JSON Schema. 2023core Model Context Protocol (MCP) An open protocol connecting AI apps to external tools and data. 2024core Reflexion Agents write verbal self-reflections after failures and reuse them. 2023core Generative Agents (Memory & Reflection) Agents with memory streams, reflection and planning simulate human life. 2023core Agent–Computer Interface (SWE-agent) LM-tailored command interfaces that boost automated bug fixing. 2024core Tree of Thoughts Expands reasoning as a tree searched with self-evaluation and backtracking. 2023core Multi-Agent Conversation (AutoGen) Multiple specialised LLM agents converse and call tools to solve a task together. 2023core MemGPT (Hierarchical Agent Memory) LLM manages its own memory like an OS – paging between the context window and external storage. 2023core

04Must-read papers

05Courses & videos

06Hands-on: project, code & exercises

Free, official or author-published pages. Most run in the browser or on a free Colab GPU.

PROJECT

Write an MCP server that can check the weather, read local files and run web searches; connect it to any MCP client and complete 3 multi-step tasks.

  1. 1
    huggingface.co ↗Courses

    Complete the Hugging Face Agents course and earn the certificate.

  2. 2
    modelcontextprotocol.io ↗Official docs

    Write an MCP server with the official tutorial and connect it to a client.

  3. 3
    huggingface.co ↗Official docs

    Build a small agent that searches and writes code with smolagents.

07Self-test

Answer in your head, then open to check.

What three parts make up a ReAct loop?

Thought, action, observation.

What message protocol is MCP built on?

JSON-RPC 2.0.

SWE-agent’s key finding?

Designing a dedicated agent–computer interface (ACI) for the model greatly improves results.

On these routes: Everyday Users · Application Developers · Product Managers & Leaders · Domain Scientists

Data updated: AI algorithms & curriculum(UTC+8)