Books › Edge Machine Learning Engineer roadmap

Best books for Edge Machine Learning Engineers

This list is about running machine learning on hardware with little memory, power and time. It starts with the embedded constraints that define the work, moves to the training and efficiency books that decide what can be made to fit, and ends with the product decisions about where inference runs and how models change in the field. Nothing here teaches one chip or one runtime; those change faster than the principles. The books assume you can train a small model and are willing to measure it on a device.

8 books in 3 stages, about 3,930 pages (at least 98 hours of reading). The notes are ours, with no ratings or sales claims.

See the full Edge Machine Learning Engineer roadmap →Best courses for Edge Machine Learning Engineers →

How to choose your books

A short method, so you read the right book for where you are instead of the longest list.

Find your stage

  • Getting started · 2 books

    Embedded constraints and microcontroller machine learning. Read these first to learn what a device can afford in memory, time and power.

    Jump to this stage ↓
  • Doing the job · 5 books

    Training, efficiency and the hardware trade-offs behind quantization and pruning. Read these while you convert and benchmark your first models.

    Jump to this stage ↓
  • Senior and leading · 1 books

    Product decisions for embedded AI. Read this when you decide what runs on the device, how models reach the field, and how success is measured there.

    Jump to this stage ↓

If you are unsure, start one stage lower than you think. Books in a later stage assume the working knowledge the earlier ones give you.

Begin with one book

Our starting pick

TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers

by Pete Warden, Daniel Situnayake · 2020

It shows what machine learning looks like under real memory limits, which is the constraint every later decision in this role is about.

See the book ↓

Then pick by your situation

Choose the next book by the problem in front of you, not by its place in the list.

Set a time budget

StageBooksPagesReadingAt 3 hrs a week
Getting started283021 hrs7 wks
Doing the job53,10078 hrs26 wks
Senior and leading1n/an/an/a
All 3 stages8≥ 3,930≥ 98 hrs≥ 33 wks

Treat this as a ceiling, not a target. Read one book at a time, apply it, then move on; read reference-style books in sections. Totals are approximate.

Judge any book before you buy it

  • Age

    This list runs from 2011 (Making Embedded Systems) to 2023 (AI at the Edge). Older books earn their place by teaching principles that do not change. For tools and platforms, buy the latest edition.

  • Level

    Check the stage before the title. A book from a later stage will make sense only once you have faced the problems it solves.

  • Format

    The longest book here, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, runs about 860 pages. Read very long books as references: pick the chapter that matches the problem you have.

  • Fit

    Ask what you will do differently the week after you finish. If you cannot answer, the book is for a different moment. Come back to it when the situation arrives.

Getting started

Embedded constraints and microcontroller machine learning. Read these first to learn what a device can afford in memory, time and power.

  • Start here

    TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers

    by Pete Warden, Daniel Situnayake · 2020

    ~500 pages~13 hrs to read

    Running machine learning on microcontrollers with a few kilobytes of memory. It is the clearest introduction to the constraints that define edge work.

    View book →
  • Making Embedded Systems: Design Patterns for Great Software

    by Elecia White · 2011

    ~330 pages~8 hrs to read

    How embedded software is structured: real-time constraints, hardware interfaces and testing on target. Read it before you integrate a model into a device's firmware.

    View book →

Doing the job

Training, efficiency and the hardware trade-offs behind quantization and pruning. Read these while you convert and benchmark your first models.

  • Embedded Systems Architecture: A Comprehensive Guide for Engineers and Programmers

    by Tammy Noergaard · 2013

    ~670 pages~17 hrs to read

    The hardware and software layers of an embedded product, from processors and memory to operating systems and drivers. It helps you read a datasheet and understand what a runtime can use.

    View book →
  • Cover of Deep Learning with PyTorch

    Deep Learning with PyTorch

    by Eli Stevens, Luca Antiga, Thomas Viehmann · 2020

    ~520 pages~13 hrs to read

    Training and modifying neural networks in PyTorch, which is the usual starting point before a model is exported to an edge runtime. Read the training chapters before you quantize anything.

    View book →
  • Cover of Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow

    Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow

    by Aurélien Géron · 2019

    ~860 pages~22 hrs to read

    A practical path through model training and evaluation with TensorFlow and Keras. The evaluation chapters are the base for the accuracy regression tests you will need on device.

    View book →
  • Efficient Processing of Deep Neural Networks

    by Vivienne Sze, Yu-Hsin Chen, Tien-Ju Yang, Joel Emer · 2020

    ~430 pages~11 hrs to read

    How the cost of a network depends on data movement, precision and hardware design. It explains why quantization and pruning help, and where they do not.

    View book →
  • Deep Learning for Coders with fastai and PyTorch: AI Applications Without a PhD

    by Jeremy Howard, Sylvain Gugger · 2020

    ~620 pages~16 hrs to read

    A top-down course in building and training models that produces working code early. Use it to get a trained baseline to convert, since edge work starts from a model that already works.

    View book →

Senior and leading

Product decisions for embedded AI. Read this when you decide what runs on the device, how models reach the field, and how success is measured there.

  • AI at the Edge: Solving Real-World Problems with Embedded Machine Learning

    by Daniel Situnayake, Jenny Plunkett · 2023

    Product-level decisions for embedded AI: where inference runs, how models are updated and how a team measures success on a device. Read it when you are deciding what runs locally and what does not.

    View book →

Book links go to Amazon and are affiliate links. As an Amazon Associate I earn from qualifying purchases.

Frequently asked questions

Which book should a Edge Machine Learning Engineer read first?

TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers by Pete Warden, Daniel Situnayake. It shows what machine learning looks like under real memory limits, which is the constraint every later decision in this role is about.

Do I need to read all 8 books?

No. The list is a menu, not a syllabus. Find your stage (getting started: 2, doing the job: 5, senior and leading: 1), begin with one book (we suggest TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers), then pick the next book by the problem you are facing at work. Come back to the others when the situation they cover arrives.

How long does it take to read these books?

The 7 books with a known page count add up to about 3,930 pages, roughly 98 hours at 1.5 minutes a page. At three hours a week that is about 33 weeks; at six hours a week, about 16. The list is meant to be read over time, one book at a time, not in one run.

Are these books up to date?

They range from 2011 (Making Embedded Systems) to 2023 (AI at the Edge). Older titles are included because they teach principles that have not changed; for tools and platforms, we favor recent editions. Check the edition before you buy.

Some links are affiliate links: we earn a small commission if you buy, at no cost to you. It does not change what we list.