What problem it solves

Machine learning models must operate within practical constraints such as memory, latency, power, reliability, and safety. Existing books and papers often cover models or individual systems separately, leaving a need for material that teaches the infrastructure of machine learning systems as one subject.

Key features

• Four-volume systems curriculum — The series progresses from single-machine execution to distributed clusters, autonomous agents, and physical AI systems.

  Source: README, books

• Hands-on framework building — TinyTorch guides students through building a machine learning framework from tensors to a transformer across 20 progressive modules.

  Source: README, tinytorch

• Interactive systems labs — Browser-based Marimo notebooks pair with textbook chapters and let students predict outcomes, change parameters, and observe what breaks.

  Source: README, labs

• System modeling and hardware practice — MLSys·im models memory, compute, network, and energy limits, while hardware kits deploy models to devices subject to real resource constraints.

  Source: README, mlsysim

• Teaching and practice resources — The repository includes course syllabi, weekly readings, lecture slides, systems design questions, practice drills, and mock interviews.

  Source: README, instructors

Who it may suit

Inference: The materials may suit students, self-learners, instructors, and contributors who want to study or teach machine learning systems from foundational principles through practical implementation.

Topics to discuss

• Which additional systems constraints should be integrated into future editions of the curriculum?

• How can the textbook, TinyTorch, MLSys·im, labs, and hardware kits form a more coherent learning sequence?

• Which examples would best connect distributed machine learning systems with agentic and physical AI?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/harvard-edge/cs249r_book#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.