🤖 Machine Learning Track¶
From linear regression to deploying models at scale. Math-first, framework-second.
🎯 Target Roles¶
- Machine Learning Engineer
- Data Scientist
- Computer Vision Engineer
- NLP Engineer
- MLOps Engineer
- Research Scientist (ML/AI)
📚 Core Courses¶
Foundations (Math for ML)¶
| Course | Platform | Institution | Level |
|---|---|---|---|
| Mathematics for Machine Learning | Coursera (audit) | Imperial College | Beginner |
| Linear Algebra | MIT OCW | MIT (Gilbert Strang) | Beginner |
| Probability and Statistics | NPTEL | IIT Kanpur | Intermediate |
| Essence of Linear Algebra | YouTube | 3Blue1Brown | Beginner |
| Multivariate Calculus | YouTube | 3Blue1Brown | Beginner |
| Statistical Learning | edX (audit) | Stanford | Intermediate |
Machine Learning¶
| Course | Platform | Institution | Level |
|---|---|---|---|
| Machine Learning (CS229) | YouTube | Stanford (Andrew Ng) | Intermediate |
| Machine Learning | Coursera (audit) | Stanford (Andrew Ng) | Beginner |
| Machine Learning | NPTEL | IIT Madras | Intermediate |
| Pattern Recognition and ML | NPTEL | IISc Bangalore | Advanced |
| Learning from Data (Caltech) | Caltech OCW | Caltech | Intermediate |
| Machine Learning (Bloomberg) | YouTube | Bloomberg | Intermediate |
Deep Learning¶
| Course | Platform | Institution | Level |
|---|---|---|---|
| Deep Learning Specialization | Coursera (audit) | deeplearning.ai | Intermediate |
| Deep Learning (CS231n) | YouTube | Stanford | Intermediate |
| Deep Learning | NPTEL | IIT Madras | Intermediate |
| Practical Deep Learning (fast.ai) | fast.ai (free) | fast.ai | Beginner |
| Neural Networks: Zero to Hero | YouTube | Andrej Karpathy | Intermediate |
| MIT 6.S191: Intro to Deep Learning | YouTube | MIT | Beginner |
Natural Language Processing¶
| Course | Platform | Institution | Level |
|---|---|---|---|
| NLP with Deep Learning (CS224N) | YouTube | Stanford | Advanced |
| Natural Language Processing | NPTEL | IIT Kharagpur | Intermediate |
| Hugging Face NLP Course | Hugging Face (free) | Hugging Face | Intermediate |
Computer Vision¶
| Course | Platform | Institution | Level |
|---|---|---|---|
| Computer Vision (CS231n) | YouTube | Stanford | Intermediate |
| Computer Vision | NPTEL | IIT Kanpur | Intermediate |
| First Principles of Computer Vision | YouTube | Columbia | Intermediate |
Reinforcement Learning¶
| Course | Platform | Institution | Level |
|---|---|---|---|
| Reinforcement Learning (CS285) | YouTube | UC Berkeley | Advanced |
| RL Course | YouTube | DeepMind (David Silver) | Intermediate |
| Reinforcement Learning | NPTEL | IIT Madras | Intermediate |
MLOps & Deployment¶
| Course | Platform | Level |
|---|---|---|
| MLOps Specialization | Coursera (audit) | Intermediate |
| Full Stack Deep Learning | Free Course | Intermediate |
| Made With ML | Free Course | Intermediate |
📖 Essential Reading (Free)¶
| Book | Topic | Link |
|---|---|---|
| Deep Learning | Theory + practice | deeplearningbook.org |
| Dive into Deep Learning | Interactive (PyTorch/TF) | d2l.ai |
| Pattern Recognition (Bishop) | Classical ML | microsoft.com |
| Speech and Language Processing | NLP | web.stanford.edu/~jurafsky/slp3 |
| Probabilistic ML | Bayesian methods | probml.github.io |
📖 Learning Path¶
Beginner: Linear algebra + Probability → Python + NumPy → Classical ML (sklearn)
Intermediate: Neural networks → CNNs/RNNs → PyTorch/TensorFlow → One domain (CV/NLP/RL)
Advanced: Transformers → Large-scale training → MLOps → Research papers
Expert: Novel architectures → Publishing → Production ML systems at scale
📓 Short Courses & Hands-On (From Learning Log)¶
| Course | Platform | Link |
|---|---|---|
| Quantization Fundamentals with Hugging Face | DeepLearning.AI | deeplearning.ai |
| Building Multimodal Search and RAG | DeepLearning.AI | deeplearning.ai |
| Build Apps with Windsurf's AI Coding Agents | DeepLearning.AI | deeplearning.ai |
| MLOps Concepts | Datacamp | Datacamp |
| Machine Learning Monitoring Concepts | Datacamp | Datacamp |
Books¶
| Book | Link |
|---|---|
| Designing Machine Learning Systems (Chip Huyen) | O'Reilly |
| Grokking Machine Learning | Manning |
"All models are wrong, but some are useful." George Box