Course curriculum
Stage 1
Pre-Foundations
- 1120 min
The ML Engineering Landscape
- 2120 min
Math & Programming Refresher
- 3
The production ML engineering curriculum for 2025-2026 — built for engineers who need to ship models that hold up under real load. You go from PyTorch 2.4+ fundamentals through distributed training on H100/H200 GPUs (DeepSpeed ZeRO-3, FSDP), MLOps (MLflow 2.16+, vLLM 0.6+, Feast 0.40+), and the LLM stack (LoRA arxiv 2106.09685, FlashAttention-2 arxiv 2307.08691, Mixtral arxiv 2401.04088). What is unique: every module ends with an interview simulation built on the question patterns Stripe, Netflix, Uber, Pinterest, and Anthropic use in their senior MLE rounds, plus a one-page runbook artifact you can show in a portfolio.
The ML Engineering Landscape
Math & Programming Refresher
Setting Up Your ML Environment
Classical Machine Learning
Feature Engineering & Data Pipelines
Deep Learning Fundamentals
Experiment Tracking & Reproducibility
Natural Language Processing
Computer Vision Engineering
MLOps & Model Deployment
Model Optimization & Efficiency
Large Language Model Engineering
Distributed Training & Large-Scale ML
ML System Design
Responsible AI & ML Ethics
Interview Preparation for ML Engineers
Generative AI & Multimodal Systems
ML Platform Engineering
Edge ML & On-Device Intelligence
Adversarial Robustness and ML Security
Data Drift and Retraining Strategies
Continual Learning and Online Learning
ML Cost Optimization and FinOps
AutoML and Neural Architecture Search
ML Theory Round Prep
ML Coding & Algorithms (Interview)
ML System Design (Interview Prep)
Applied ML & Productionization