A cram sheet for ML engineering interviews, by someone who survived them

What it does This repo is a curated reading list and study schedule for machine learning engineering interviews. It bundles links to LeetCode categories, SQL practice, probability cheat sheets, and system design scenarios into a single index. The author—a software engineer with ten years of ML experience—also peppers in links to their own books and interview stories.

The interesting bit Unlike most prep guides that demand you learn everything, this one tells you what you can safely ignore. It explicitly warns that Spark and Cassandra knowledge is “NOT required for Google, Facebook interview” and admits that plenty of companies skip LeetCode entirely. That kind of triage is usually missing from interview lore.

Key highlights

• Six ML system design walkthroughs pulled from real products: YouTube recommendations, LinkedIn feed ranking, Airbnb search, ad click prediction, and delivery time estimation.
• A LeetCode tracking spreadsheet and categorized question lists for candidates who need to grind.
• “Implement from scratch” prompts for logistic regression and K-means, with bonus points for vectorized NumPy solutions completed in twenty minutes.
• A probability cheat sheet and links to core ML concepts: collinearity, SMOTE, L1/L2 regularization, and boosting.

Caveats

• The repository is mostly a link directory; original code and prose live elsewhere.
• Several sections double as marketing funnels for paid books and courses on Amazon, Educative, and the author’s blog.
• Some big-data notes are flagged as outdated, specifically pre-Spark 3.0 architecture guidance.

Verdict Bookmark it if you want a battle-tested syllabus for FAANG-style ML engineering loops. Skip it if you need a self-contained course or original code to run; this is a table of contents, not the classroom.