Aurélien Géron's practical tour from linear models to deep learning.
Why I recommend itThe book I recommend most to people starting out — every chapter ends in runnable code.
View on Amazonon amazon.comToolkit
The resources I actually recommend
After 10 years of building ML in production and mentoring data scientists, these are the books, courses and tools I point people to. Short list, no fluff.
Chip Huyen on taking models from notebook to production.
Why I recommend itThe best single read on MLOps, data drift and monitoring that I know of.
View on Amazonon amazon.comBooks
AI Engineering
Chip Huyen on building applications with foundation models.
Why I recommend itCovers RAG, evaluation and fine-tuning trade-offs without the hype.
View on Amazonon amazon.comFrançois Chollet (creator of Keras) explains deep learning from first principles.
Why I recommend itBuilds real intuition for why neural networks work, not just how to call them.
View on Amazonon amazon.comThe 50+ statistical concepts data scientists actually use.
Why I recommend itPerfect desk reference for hypothesis tests, A/B testing and resampling.
View on Amazonon amazon.comStanford × DeepLearning.AI's beginner-friendly ML course on Coursera.
Why I recommend itStill the gentlest, clearest on-ramp into ML I've seen.
Start learningon coursera.orgCourses
Deep Learning Specialization
Five courses on neural networks, CNNs, sequence models and transformers.
Why I recommend itGreat follow-up once you know the basics and want the math behind deep learning.
Start learningon coursera.orgCourses
DataCamp
Interactive, in-browser courses for Python, SQL and data science.
Why I recommend itBest for learning by typing code daily in short sessions.
Try DataCampon datacamp.comTools
Weights & Biases
Experiment tracking, model registry and LLM evaluation.
Why I recommend itWhat I use to stop losing track of which run produced which model.
Try it freeon wandb.aiTools
Google Colab Pro
Hosted Jupyter notebooks with faster GPUs.
Why I recommend itThe cheapest way to get a GPU for training experiments without any setup.
See planson colab.research.google.comAI assistants
ChatGPT
OpenAI's assistant for writing, coding, analysis and brainstorming.
Why I recommend itMy go-to for quick drafts and talking through a modelling idea out loud.
Try ChatGPTon chatgpt.com · Not sponsoredAI assistants
Claude
Anthropic's assistant, strong at long documents, careful reasoning and code.
Why I recommend itWhat I reach for when I need to read a long paper or refactor a messy notebook.
Try Claudeon claude.ai · Not sponsoredAI assistants
Google Gemini
Google's assistant, built into Search, Gmail, Docs and Android.
Why I recommend itHandy when the work already lives in Google Docs, Sheets or Drive.
Try Geminion gemini.google.com · Not sponsoredSome links are affiliate links: if you buy through them I may earn a small commission, at no extra cost to you. I only recommend things I've used or would recommend to my own team. Full disclosure