# harvard-edge/cs249r_book

Machine Learning Systems

Repository: https://github.com/harvard-edge/cs249r_book
Canonical: https://ross.abutalabs.com/products/cs249r_book
Homepage: http://mlsysbook.ai/
Language: Python
License: NOASSERTION
License Family: other
Topics: artificial-intelligence, cloud-ml, computer-systems, courseware, deep-learning, edge-machine-learning, embedded-ml, machine-learning, machine-learning-systems, mobile-ml, textbook, tinyml
Last push: 2026-08-27T00:30:27+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 90, longevity 78
- inputs: {"age_days": 1092, "days_push": 7, "days_rel": 70, "gap_med": 0, "n_releases_24m": 26}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 28010, forks 3535 (observed 2026-08-28T04:11:48.042977+00:00)

## What it is
An open-access, two-volume textbook and curriculum ecosystem on engineering machine learning systems, from single-device foundations to fleet-scale infrastructure, published by Harvard with MIT Press. It bundles interactive labs, TinyTorch (a build-your-own-framework course), hardware kits for embedded deployment, performance modeling tools, and instructor materials.

## Use cases
- learn how ML systems are built and optimized end to end
- study machine learning systems engineering as a course
- build a deep learning framework from scratch to understand internals
- deploy machine learning models on embedded hardware like Arduino and Raspberry Pi
- model GPU performance bottlenecks for LLM inference
- prepare for ML systems engineering interviews
- adopt an open curriculum for teaching AI engineering at a university

## When to choose
- you want a rigorous, principles-first treatment of ML systems rather than just algorithms
- you need free, openly licensed courseware with labs, slides, and instructor materials
- you want hands-on practice deploying ML to constrained edge hardware
- you prefer learning framework internals by building TinyTorch yourself

## When to avoid
- you need a production ML framework or library rather than educational material
- you want only ML theory or algorithm tutorials without systems/hardware context
- you need commercial-use rights to the textbook content (it is CC-BY-NC-SA)

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, deep-learning, developer-tools, documentation
- domain: machine-learning, artificial-intelligence, education, tutorials, embedded-systems
- platform: cross-platform, python
- tags: textbook, tinyml, edge-ml, ml-systems, courseware, tinytorch, open-access, hardware-kits, interactive-labs, web

## Member repositories
- harvard-edge/cs249r_book (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:48.042977+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 2026-08-29T16:54:28.612832+00:00, confidence not recorded.
  - readme: https://github.com/harvard-edge/cs249r_book (fetched 2026-08-28T04:11:48.042977+00:00, sha b321a4269ec3)
  - homepage: http://mlsysbook.ai/ (fetched 2026-08-29T07:51:19.947048+00:00, sha 39104d4ed4fc)
  - site_page: http://mlsysbook.ai/about/index.html (fetched 2026-08-29T07:51:19.951051+00:00, sha 2534af2a32fb)
  - site_page: http://mlsysbook.ai/about/people.html (fetched 2026-08-29T07:51:19.953272+00:00, sha 370da3f3637b)
  - site_page: http://mlsysbook.ai/about/contributors.html (fetched 2026-08-29T07:51:19.954996+00:00, sha e41d32ae445a)
  - site_page: https://mlsysbook.ai/about (fetched 2026-08-29T07:51:19.956504+00:00, sha df80a0c6901b)
  - site_page: https://mlsysbook.ai/about/people.html (fetched 2026-08-29T07:51:19.958327+00:00, sha 370da3f3637b)
  - site_page: https://mlsysbook.ai/about/contributors.html (fetched 2026-08-29T07:51:19.959824+00:00, sha e41d32ae445a)
  - site_page: https://mlsysbook.ai/about/license.html (fetched 2026-08-29T07:51:19.961243+00:00, sha 5da18bb138e0)
- Data as of 2026-08-30T08:39:29.467469+00:00.