Codebase Tutorials

AI-generated, chapter-by-chapter walkthroughs of real open-source codebases. Each tutorial breaks a project into its core abstractions and explains them from the ground up.

73 tutorials
9
99
8 chapters

The 99 project is a sophisticated AI assistant plugin for Neovim that bridges your code editor with powerful AI models. It acts as a "Control Tower," managing context-aware Requests that leverage LSP…

A
Agent-Skills-for-Context-Engineering
7 chapters

This project is a comprehensive library of Agent Skills designed to master Context Engineering—the discipline of managing an AI model's limited attention and memory budget. By utilizing concepts like…

A
AionUi
7 chapters

AionUi is a sophisticated desktop AI workspace that orchestrates complex tasks across multiple AI backends (like Gemini, Codex, and ACP) through a unified interface. It acts as a central hub where Ag…

A
aios-core
7 chapters

AIOS-Core is a comprehensive framework designed to treat AI agents as a cohesive, professional software development team. It utilizes a Master Orchestrator to manage complex project lifecycles ("Epic…

A
airi
7 chapters

airi is a modular ecosystem designed to create living digital characters that can inhabit various platforms, from Minecraft bots to desktop companions. At its core is a Cognitive Brain that uses Larg…

A
antigravity-awesome-skills
5 chapters

Antigravity Awesome Skills acts as a comprehensive operating system for AI agents, providing a massive library of atomic Agentic Skills that transform generalist models into specialized experts (like…

A
autogen
7 chapters

Autogen is a framework designed to build autonomous Agents that can collaborate to solve tasks. It provides a robust Runtime layer to manage agent lifecycles and route Messages, allowing agents to co…

A
autogen
7 chapters

AutoGen is a framework for building multi-agent AI systems where Agents act as autonomous units that can converse, execute code, and interact with LLMs. These agents are organized into Teams to colla…

A
awesome-generative-ai-guide
3 chapters

This project serves as a comprehensive Generative AI resource hub, offering curated lists of research papers, free courses, and interview preparation materials. It tracks the rapid evolution of the f…

B
botan
4 chapters

Botan is a comprehensive C++ cryptography library designed to provide the tools necessary for implementing practical systems, such as TLS, certificates, and modern ciphers. It focuses on performance…

C
C-Plus-Plus
9 chapters

This project is a versatile collection of C++ implementations that spans core computer science domains. It provides efficient examples of fundamental data structures and algorithms, such as sorting a…

C
cc-switch
6 chapters

cc-switch acts as a smart Local Proxy Gateway that sits between your AI command-line tools (like Claude Code) and various AI providers. Instead of connecting directly to the internet, tools connect t…

C
chrome-devtools-mcp
6 chapters

This project serves as a Model Context Protocol (MCP) server that bridges the gap between an AI model and a Chrome browser. It allows the AI to act as a developer by inspecting the page's accessibili…

C
claude-pilot
7 chapters

Claude Pilot is a comprehensive wrapper for the Claude Code CLI that transforms it into a persistent, memory-aware developer assistant. It uses a Lifecycle Hooks System to intercept CLI events and fo…

C
claudeCode
19 chapters · 35 subsystems

claudeCode is an advanced AI CLI tool designed to act as an autonomous software engineer. It features a robust Query Engine that orchestrates conversation loops, manages state, and executes a diverse…

C
ClickHouse
14 chapters

ClickHouse is a high-performance OLAP database designed for real-time analytics. The project includes a sophisticated CI/CD infrastructure powered by the Praktika framework to orchestrate builds and…

C
compound-engineering-plugin
7 chapters

The Compound Engineering Plugin is a comprehensive toolkit for AI-assisted software development that enforces a Plan → Work → Review → Compound cycle to ensure engineering work becomes easier over ti…

C
crewAI
9 chapters

crewAI is a framework designed for orchestrating AI agents, enabling them to work together using various LLM providers like OpenAI, Azure, and Anthropic. The provided code configures the project's te…

C
crewAI
6 chapters

CrewAI is a framework for orchestrating autonomous AI agents that collaborate as Crews to complete complex objectives. It allows developers to build efficient, production-ready systems by combining f…

C
cs249r_book
12 chapters

TinyTorch is an educational framework for building Machine Learning Systems from scratch, specifically focusing on the implementation of Transformer architectures. It enables users to construct, trai…

C
cutlass
16 chapters

CUTLASS is a high-performance CUDA C++ template library for implementing matrix-matrix multiplication (GEMM) and related linear algebra primitives. It provides hierarchical abstractions to efficientl…

D
deer-flow
7 chapters

deer-flow is a comprehensive AI agent platform that orchestrates a Lead Agent and specialized Sub-Agents to handle complex, multi-step tasks. It utilizes a secure Sandboxed Environment for safe code…

D
dexter
7 chapters

Dexter is an autonomous financial research agent that bridges natural language queries with complex financial data. Instead of a linear script, it uses a recursive loop to iteratively "think," select…

D
dozzle
7 chapters

Dozzle is a real-time log viewer for Docker containers that provides a modern, responsive web interface. It acts as a Universal Translator for container metrics and logs, supporting local Docker inst…

E
entireio-cli
6 chapters

Entire is a CLI tool that bridges the gap between AI coding assistants (like Claude Code or Gemini) and Git. It automatically captures conversation history and code changes as checkpoints, managing t…

F
fastapi
7 chapters

FastAPI is a modern, high-performance web framework for building APIs with Python. It acts as a central App Instance that orchestrates request handling, leveraging standard Python type hints to perfo…

F
filebrowser
7 chapters

This project is a web-based file manager that allows users to manage, browse, and share files on a server through a modern interface. It acts as a virtual librarian, maintaining an index of the physi…

G
garak
7 chapters

garak serves as an automated vulnerability scanner for Large Language Models (LLMs), functioning similarly to a red team or penetration tester. It orchestrates a workflow where Probes generate malici…

G
gastown
7 chapters

Gas Town is a multi-agent orchestration system designed to manage AI coding workflows with persistent state. It organizes work into self-contained Rigs, employs ephemeral Polecat agents to execute ta…

G
get-shit-done
9 chapters

get-shit-done is an autonomous AI project management framework that uses high-level Orchestrators to drive the software development lifecycle. It persists context in a Project State memory to prevent…

G
GitHub-Agentic-Workflows
7 chapters

GitHub Agentic Workflows is a secure framework that compiles high-level Markdown workflow definitions into executable GitHub Actions. It acts as an orchestrator for AI agents, wrapping them in a rigo…

G
GitNexus
7 chapters

GitNexus is a developer tool that transforms a raw codebase into a structured Knowledge Graph, enabling AI agents to understand execution flows, dependencies, and the impact of changes. It functions…

G
gogcli
6 chapters

gogcli is a robust command-line interface designed to interact with Google Workspace services like Gmail and Drive. It employs a structured command framework to separate argument parsing from busines…

H
Hands-On-AI-Engineering
7 chapters

This project builds intelligent digital employees that go beyond simple chatbots. It uses specialized AI Agents that work together in teams (orchestration) to solve complex tasks like financial analy…

H
Hands-On-Large-Language-Models
7 chapters

This project serves as a comprehensive, hands-on guide to working with Large Language Models (LLMs). It covers the full lifecycle of LLM application development, from creating generative pipelines an…

J
jarvis-ai-assistant
6 chapters

A robust desktop AI assistant built with Electron that bridges native system inputs (like global hotkeys and microphone access) with a powerful hybrid transcription engine capable of switching betwee…

L
langchain
7 chapters

LangChain is a framework for building applications powered by Language Models (LLMs and Chat Models). It provides a structure where Runnables & Chains orchestrate sequences of calls, Agents use model…

L
langextract
7 chapters

LangExtract is a framework designed to turn unstructured text into structured data (like JSON or YAML) using Large Language Models (LLMs). It orchestrates the entire pipeline by breaking long documen…

L
LLMs-from-scratch
8 chapters

This project serves as a comprehensive educational guide for building Large Language Models (LLMs) from the ground up. It demonstrates how to transform raw text into numerical data using a Tokenizer,…

M
Made-With-ML
8 chapters

This project is an end-to-end MLOps platform designed to classify machine learning projects using a Large Language Model (BERT). It demonstrates the complete lifecycle of an AI application, from proc…

M
markitdown
7 chapters

MarkItDown is a versatile utility designed to convert a wide variety of files and web content into clean, readable Markdown. Acting as a universal translator, it automatically detects the input type—…

M
mcp-interviewer
8 chapters

The mcp-interviewer is an automated evaluation framework for Model Context Protocol (MCP) servers. Acting as a digital hiring manager, it orchestrates a complete lifecycle that includes connecting to…

M
Megatron-LM
9 chapters

Megatron-LM is a high-performance deep learning library designed by NVIDIA for training massive scale artificial intelligence models. It enables researchers to efficiently train huge text generators…

M
memU
7 chapters

memU is a framework that provides AI agents with long-term memory by organizing information into a hierarchical structure (Resources, Items, Categories) rather than a simple list of vectors. Acting a…

M
ML-For-Beginners
15 chapters

This project is a comprehensive 12-week curriculum designed to introduce Machine Learning concepts to beginners. It features a mix of written lessons, code demonstrations in Python and R, and an inte…

M
ML-For-Beginners
18 chapters

This project is an educational resource for learning Machine Learning, specifically focusing on Natural Language Processing (NLP). It features a Romanian translation of a lesson that teaches how to p…

M
moonshine
7 chapters

Moonshine is a lightweight, fast speech-to-text library designed for real-time applications on edge devices and the cloud. At its core, a Transcriber orchestrates the flow of audio through Voice Acti…

M
multi-agent-marketplace
6 chapters

This project simulates a digital economy where autonomous AI agents (customers and businesses) interact to trade goods and services. It uses a central platform to enforce strict market protocols, ens…

N
nanobot
6 chapters

Nanobot is a modular AI assistant designed to bridge external chat platforms (like Telegram and WhatsApp) with powerful Large Language Models. Its core "brain," the Agent Loop, orchestrates conversat…

N
nautilus_trader
4 chapters

Nautilus Trader is a high-performance, event-driven platform designed for algorithmic trading and backtesting. It features a modular architecture written in Rust, utilizing various adapters to connec…

N
NeMo-Agent-Toolkit
7 chapters

The NeMo Agent Toolkit is a modular framework designed for building, deploying, and evaluating intelligent Generative AI agents. It features a central Workflow Builder to assemble components like LLM…

N
novel
7 chapters

Novel is a Notion-style WYSIWYG editor framework built on top of Tiptap and React. It utilizes a headless architecture to decouple the core editing engine from the user interface, allowing developers…

O
open-notebook
7 chapters

Open Notebook is an intelligent research assistant designed to help users ingest, organize, and chat with their documents. It leverages a Content Processing Pipeline to convert raw files (PDFs, text)…

O
openclaw
10 chapters

OpenClaw is a modular AI agent platform that unifies various devices into a single intelligent network. At its core is a Gateway daemon that manages connections, executes multi-agent workflows define…

P
pi-mono
7 chapters

pi-mono is an advanced AI coding assistant built on a modular architecture. At its core, an Agent Runtime manages the thought loop, while an Agent Session handles long-term memory and persistence. It…

P
picoclaw
7 chapters

PicoClaw is an ultra-lightweight personal AI agent designed to run on embedded devices and development boards. It acts as a central "brain" that connects various communication channels (like Telegram…

P
PQC
6 chapters

This project implements a Post-Quantum Cryptography (PQC) enabled microservices architecture using gRPC and C++. It specifically utilizes OpenSSL 3.5 to upgrade standard TLS connections to use hybrid…

P
Prompt-Engineering-Guide
13 chapters

The Prompt Engineering Guide is a comprehensive, open-source educational resource designed to help developers and researchers master interactions with Large Language Models (LLMs). It organizes a vas…

P
pydantic
7 chapters

Pydantic is a data validation and settings management library that uses Python type annotations to define data schemas. At its core, users define models inheriting from BaseModel with typed Fields an…

Q
qmd
6 chapters

QMD is an on-device search engine designed to index and retrieve knowledge from markdown notes, documentation, and transcripts using a hybrid approach. It combines keyword search (BM25), vector seman…

R
remotion
8 chapters

Remotion is a toolkit that allows developers to create videos programmatically using React. Instead of using traditional video editing software, users write code to define sequences, compositions, an…

R
rowboat
6 chapters

Rowboat is an open-source, local-first AI coworker that turns your file system into a structured Knowledge Graph stored in plain Markdown. It features an intelligent Agent Runtime that executes tasks…

S
scikit-learn
13 chapters

Scikit-learn is a comprehensive Python library designed for machine learning and data analysis. It provides a unified interface to apply various algorithms—such as Linear Models, Trees, and Clusterin…

S
Scrapling
6 chapters

Scrapling is a high-performance Python library designed to make web scraping undetectable and resilient. It features a unique Adaptive Parser that uses fuzzy matching to intelligently relocate elemen…

S
seerr
7 chapters

Seerr is a media discovery and request management application that integrates with existing media servers like Plex, Jellyfin, and Emby. It allows users to browse and request movies and TV shows thro…

S
shannon
9 chapters

This project functions as an autonomous cybersecurity agent designed to simulate cyber attacks for security testing. It systematically gathers intelligence from various data sources, formulates a str…

S
squad
6 chapters

Squad transforms ephemeral AI chats into a persistent, file-based development team living directly in your repository. It manages AI agents with specific roles (like Frontend or Tester) that possess…

S
superpowers
6 chapters

Superpowers is a comprehensive workflow engine for AI coding agents that transforms standard software engineering best practices into executable "skills" (markdown files). It separates the developmen…

S
system-design-primer
5 chapters

This project serves as a comprehensive educational guide for mastering system design and object-oriented design interviews. It provides architectural blueprints for large-scale services like Twitter…

V
velero
7 chapters

Velero is a cloud-native tool designed to backup and restore Kubernetes cluster resources and persistent volumes. It enables disaster recovery, data migration, and the replication of production clust…

V
vitefu
5 chapters

This project acts as an intelligent dependency crawler for Vite that automatically detects and configures "framework" packages (libraries containing raw, non-standard code). It traverses the dependen…

W
wifi-densepose
7 chapters

A WiFi-based perception system that converts invisible WiFi signal reflections (CSI) into human pose estimates and vital sign analytics. The project features a dual-mode architecture: a DensePose pip…

Z
zvec
5 chapters

Zvec is an embedded vector database engine designed to bridge high-performance C++ internals with user-friendly Python bindings. It organizes data into manageable chunks called segments and employs a…