What is ai-engineering-from-scratch built with?

Q: Which repos have a stack like ai-engineering-from-scratch's?

The ones that share most of its stack: comet-ml/opik, datawhalechina/hello-agents, and unslothai/unsloth.

ai-engineering-from-scratch is a Python repo built with Hono. It uses AI through accelerate, librosa, tokenizers, torchaudio, transformers, and 11 more, makes its content with pillow and shiki, validates its data with Zod, and deploys with @hono/node-server. It has 506,014 written lines, with 781 test files.

Read on November 4, 2026, at commit 8fec9cc. Read it as markdown

Frontend

Interface 1

matplotlib3 files

Backend

AI 14

torch62 files
scikit-learn8 files
torchvision4 files
mcp2 files
@mastra/core1 file
google-adk1 file
LangChain1 file
accelerate
librosa
tokenizers
torchaudio
tiktoken3 files

2 more

datasets1 file
transformers

Services 2

Support

Content 2

shiki1 file
pillow

Tools 1

Jupyter

Validation 1

Zod6 files

The base 1

Hono10 files

Deploy 1

@hono/node-server2 files

Other 5

NumPy454 files
pandas2 files
pagefind1 file
ws2 files
soundfile

Other repos with a similar stack

Ordered by how much of its stack they share with ai-engineering-from-scratch, then by stars.

comet-mlopik
23.2k
Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.
12 of 28 in commonCompare →
datawhalechinahello-agents
86k
📚 《从零开始构建智能体》——从零开始的智能体原理与实践教程
11 of 28 in commonCompare →
unslothaiunsloth
79.1k
Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.
11 of 28 in commonCompare →
debpalashVoiceStudio
100.7k
VoiceStudio is the open-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing, dictation, transcription & audiobook creation in 646 languages.
10 of 28 in commonCompare →
vllm-projectvllm
96.3k
A high-throughput and memory-efficient inference and serving engine for LLMs
10 of 28 in commonCompare →
hiyougaLlamaFactory
75.9k
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
8 of 28 in commonCompare →

Frequently asked questions

What language is ai-engineering-from-scratch written in?

ai-engineering-from-scratch is written in Python, with 506,014 written lines.

What framework does ai-engineering-from-scratch use?

ai-engineering-from-scratch is built with Hono.

Does ai-engineering-from-scratch use AI?

Yes. ai-engineering-from-scratch uses AI through accelerate, librosa, tokenizers, torchaudio, transformers, torch, scikit-learn, torchvision, tiktoken, mcp, datasets, google-adk, LangChain, Anthropic, OpenAI, and @mastra/core.

Does ai-engineering-from-scratch have tests?

Yes. ai-engineering-from-scratch has 781 test files.

Which repos have a stack like ai-engineering-from-scratch's?

The ones that share most of its stack:comet-ml/opik datawhalechina/hello-agents unslothai/unsloth