AI/ML Prep
Prepare for ML engineer and applied scientist loops: theory, ML coding and ML system design.
About this course
ML interviews combine three rounds that most courses treat separately: ML fundamentals, hands-on ML coding, and ML system design. AI/ML Prep covers all three in one track.
You'll review the theory interviewers probe, from bias-variance to transformers, implement classic algorithms from scratch in NumPy, and work through five production ML design case studies, from recommendations to content moderation. A section on large language models covers the retrieval, evaluation and serving questions that now come up in most loops.
What you'll learn
- Explain bias-variance, regularisation and model evaluation clearly
- Choose the right metric for imbalanced and ranking problems
- Implement linear regression, logistic regression and k-means from scratch
- Describe how transformers and attention work at interview depth
- Design retrieval-augmented generation (RAG) systems and evaluate them
- Lead an ML system design interview from problem framing to monitoring
- Discuss training-serving skew, drift and online experimentation
Course index
9 sections · 42 lessons · 14h total · 4 free previews
- Welcome and the ML interview loopPreview10m
- ML engineer vs applied scientist vs data scientist interviewsPreview12m
Frequently asked questions
Is this for data scientists or ML engineers?
Both, with a lean toward ML engineering. Product data science loops focus more on statistics and experimentation, which the evaluation and A/B testing sections cover.
What maths background do I need?
Comfort with basic linear algebra, probability and calculus. Refreshers are included where they're needed.
Which libraries are used in the coding problems?
NumPy and pandas for coding problems, with PyTorch for the deep learning lessons.
Should I also take System Design Prep?
It helps. ML system design builds on general distributed-systems concepts such as caching, queues and storage, which System Design Prep covers in depth.
Pairs with the core courses
Most role-specific loops still include these rounds.
Data structures, algorithms and problem-solving patterns for coding rounds
143 questions · 79+ hours
Designing scalable, reliable distributed systems for architecture rounds
13 questions · 28+ hours
Storytelling, leadership and level-calibrated answers for behavioral rounds
17 questions · 15+ hours
Other add-ons
JavaScript, React, browser internals and frontend architecture interviews
17 questions · 21+ hours
Linux, CI/CD, infrastructure as code and cloud platform interviews
11 questions · 13+ hours
Swift, SwiftUI, UIKit and mobile architecture interviews
8 questions · 12+ hours
Kotlin, Jetpack Compose, coroutines and Android architecture interviews
5 questions · 11+ hours
Smart contracts, Solidity, the EVM and cryptographic protocols
8 questions · 11+ hours
Ready to start AI/ML Prep?
ML fundamentals, ML coding and large-scale ML system design