My CS Degree - Full-Stack ML Engineering (2020)

Being a machine learning software engineer with a background in Physics, I felt the lack of a CS degree had been a limiting factor in my long-term growth. To tackle it head-on, I designed this curriculum for myself with the best resources I found online, focusing on CS basics as well as full-stack development, deep learning and natural language processing.

There are general knowledge courses and project courses.

General knowledge courses are for indexing knowledge in the brain into an organized system. When facing a new problem, at least you know what relevant info to look for.

Project courses are the real learning process. Learning by doing is the only way to learn.

If you are interested in the philosophy of the creation of this curriculum, I wrote an article about it:

How I Designed My Own Full-Stack ML Engineering Degree

General knowledge courses

CS101. Computer architectures: general intro

• Coursera: From NAND to Tetris part I

CS102. Networking: general intro

• Coursera computer networking
• Stanford course youtube playlist
• parrt cs601 Sockets
• parrt cs601 Network programming
• Socket programming in Python, 5 videos

CS103. Operating systems: general intro

• Udacity course
• Coursera: From NAND to Tetris part II

CS201. Database: general intro

• Architecture of a Database System Paper
• Berkeley Course Playlist

CS401. System Design

• Book: Designing Data Intensive Applications
• Course: MIT course playlist
• Grokking the System Design Interview
• Gaurav Sen Playlist
• Uber engineering blog series

CS403. Design Patterns

• Python Factory pattern and Message Queue
• Book: Design Patterns by Gamma et al

CS405. Software Engineering Fundamentals

• Course: Software Development Process
• Prof. Terence Parr, USF CS601
• Gaurav Sen

CS406 DevOps with Docker

• Udemy Docker course
• Deploy ML/DL using Docker

CS501 Deep Learning with PyTorch by Yann LeCun & Alfredo Canziani: knowledge

• Highly recommended deep learning course!!
• Course site
• Youtube playlist

CS503 fastai: Computation Linear Algebra by Rachel Thomas

CS504. Overview of Production Machine Learning and MLOps

• Book: Building ML Powered Applications
• Suggestion: take project course CS512 for hands-on practice after this course
• Other readings:
  • MLOps overview by Huyen Chip
  • TDS Article: Big picture architecture of ML system
  • Challenges and practical guide to maintaining machine learning by Eugene Yan

CS505 Natural Language Understanding Stanford

• Course page
• Playlist

CS506 Natural Language Processing using Deep Learning CMU

CS507 Full Stack Deep Learning (Updated 2021)

• Course page
• Project template

CS508 Deep Learning for Computer Vision

CS509 Yannic Kilcher Youtube series

CS510 Luigi SageMaker course

• Great blog for prodML: mlinproduction.com

CS511 MLOps with GitHub Action

• MLOps youtube playlist
• MLOps: Setup Github Actions CI/CD for NLP project
  • Weights & Biases Youtube video with Hamel Husain
  • Blog post

CS513 Machine Learning System Design

• CS329S by Chip Huyen @Stanford
• Grokking the Machine Learning Interview

Project courses

CS301. Coding Interview

• EPI book

CS302 Python 3

• Basics: Practical Python course
• Eugene Yan's Python project setup
• RestAPI with FastAPI guide
  • FastAPI MadeWithML example
• Project: DoverChat, poetry, Docker, deployed on Heroku and AWS EB

CS303 Nature of Code youtube series

• JS Prototype
• Topics of ES6-ES8, promise, async await (20 vids)
• Nature of Code
  • Physics Engine, matter.js simulation project (10 vids)
  • Autonomous Agents (10 vids)
  • Cellular Automata (4 vids)
  • Fractals (10 vids)
    • Fractal trees, space colonization project

CS304 Weather Comparison App

• Build API using FastAPI to get daily and hourly weather for cities
• Frontend based on Streamlit, a Python framework for building UI for prototype projects
  • Streamlit get started

CS402 Full-Stack Web Development, ReactJS

• Udemy the complete junior to senior web developer roadmap

CS404 AWS Services

• Book: The Good Parts of AWS by Daniel Vassallo

CS407 iOS Development with Swift and SpriteKit

• Optional course for mobile game development
• youtube video: https://youtu.be/467Doas5J6I
• Project: A sound game for instruments

CS500 Deep Learning with FastAI & Pytorch: projects

• Course v4: https://course.fast.ai/
• fastbook
  • Document all answers to questionnaires
• FastAI2 design paper
• Deep Tutorials for PyTorch
• Official PyTorch book: Deep Learning with PyTorch

CS502 NLP: projects

• fastai NLP course
• Awesome visual course
• Book: Practical NLP
• Survey
  • Eugene Yan: NLP Survey
  • Pratik Bhavsar: Pytorch for NLP
• Other resources
  • Fast AI course v4 NLP lecture
  • @amitness: How to Learn Transformers
  • Advanced course
  • NLP Masterclass: Modeling Fallacies in NLP
  • NLP Datasets
  • Book: NLP with Pytorch
  • MadewithML and dair.ai chatbot materials
    • Survey paper
    • MadeWithML topic curation
    • Future of NLP
    • Level 3 AI Conference 2020
  • Full stack ML prototype stack: FastAPI, Streamlit, Docker, Heroku
    • Project example 1: streamlit + fastapi
    • Project example 2

CS512 Applied ML in Production by MadeWithML

• Quote the course description: "This course isn't just about ML. In fact, it's mostly about clean software engineering! We'll cover important concepts like versioning, testing, logging, etc. that really makes this a production-grade product."
• course page
• MadeWithML youtube channel

CS602 Open Source Contributions of Your Choosing