# stas00/ml-engineering

Machine Learning Engineering Open Book

Repository: https://github.com/stas00/ml-engineering
Canonical: https://ross.abutalabs.com/products/ml-engineering
Homepage: https://stasosphere.com/machine-learning/
Language: Python
License: CC-BY-SA-4.0
License Family: other
Topics: pytorch, slurm, large-language-models, llm, machine-learning, scalability, transformers, machine-learning-engineering, mlops, ai, inference, training, debugging, gpus, network, storage
Last push: 2026-08-26T00:27:36+00:00

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 100
- inputs: {"age_days": 2191, "days_push": 8, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 18739, forks 1211 (observed 2026-08-28T04:11:27.054928+00:00)

## What it is
An open-source 'open book' of methodologies, scripts, and step-by-step instructions for training, fine-tuning, and running inference on large language and multi-modal models. It distills hands-on experience from projects like BLOOM-176B and IDEFICS-80B into practical guides covering hardware, networking, storage, SLURM orchestration, training, inference, and debugging.

## Use cases
- learn how to train large language models on multi-GPU clusters
- debug NCCL and distributed training failures
- decide when to upgrade GPUs for LLM training
- set up SLURM for machine learning workloads
- optimize storage and networking for LLM training
- fine-tune and run inference on LLMs and vision-language models
- troubleshoot OOM and performance bottlenecks in PyTorch training

## When to choose
- you are an ML engineer or operator training or fine-tuning LLMs/VLMs on GPU clusters
- you need practical, copy-paste commands and battle-tested debugging recipes
- you want to understand hardware, network, and storage choices for large-scale training

## When to avoid
- you are looking for a software library or tool to install and use directly
- you are a beginner seeking introductory ML tutorials
- you need formal, peer-reviewed textbook material rather than practitioner notes

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, llm-training, llm-inference, developer-tools, documentation
- domain: large-language-models, machine-learning, deep-learning, gpu-computing, tutorials, developer-tools
- platform: python, cloud
- tags: open-book, llm-training, slurm, pytorch, distributed-training, debugging, gpu-clusters, mlops, fine-tuning, multi-modal, linux, gpu

## Member repositories
- stas00/ml-engineering (main) score 77

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:27.054928+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:01:19.101866+00:00, confidence not recorded.
  - readme: https://github.com/stas00/ml-engineering (fetched 2026-08-28T04:11:27.054928+00:00, sha 3145c2661448)
  - homepage: https://stasosphere.com/machine-learning/ (fetched 2026-08-29T07:59:17.776712+00:00, sha ce844023c293)
- Data as of 2026-08-30T08:39:29.467469+00:00.