ML / AI / LLM — Ordered Learning Path (Zero → Mastery)
This learning path takes you from mathematical foundations to production AI systems. Follow the sections in order for the best learning experience.
1. Mathematical & Data Foundations (Non-negotiable)
Build the mathematical foundation that all ML/AI is built upon.
| Topic | Description |
|---|---|
| High School Math with Python | Review fundamentals with code |
| Math Concepts Deep Dive | Essential mathematical concepts |
| Linear Algebra Projects | Vectors, matrices, transformations |
| Linear Algebra (Extended) | Advanced linear algebra topics |
| Math for Machine Learning | ML-specific mathematics |
| Statistics from Scratch | Probability and statistics |
| Data Structures from First Principles | Core data structures |
2. Programming & Data Handling for ML
Master Python and data manipulation for ML workflows.
| Topic | Description |
|---|---|
| Python Deep Dive | Master Python programming |
| Python Performance & Scaling | High-performance Python |
| Pandas Deep Dive | Data manipulation with Pandas |
| Datasets and Kaggle | Working with real datasets |
| Data Engineering Deep Dive | Data pipelines and ETL |
| Data-Intensive Applications | Designing data systems |
3. Classical Machine Learning
Traditional ML algorithms and techniques.
| Topic | Description |
|---|---|
| Machine Learning Foundations | Core ML concepts and algorithms |
| Python ML Basics | Scikit-learn and basic ML |
| String Metrics and Automata | Text similarity and matching |
| OCR Deep Dive | Optical character recognition |
| Image Processing | Computer vision basics |
4. Neural Networks & Deep Learning
Deep learning fundamentals and research.
| Topic | Description |
|---|---|
| Neural Networks Deep Dive | Neural network architectures |
| AI Research Deep Dive | Reading and implementing papers |
| ML Model Finetuning | Transfer learning and finetuning |
5. Modern LLM Foundations
Understanding large language models.
| Topic | Description |
|---|---|
| Generative AI & RAG | LLMs and retrieval-augmented generation |
| Transformers & Quantization | Transformer architecture deep dive |
| HuggingFace Ecosystem | Working with HuggingFace |
| Quantization & Inference Optimization | Model compression techniques |
6. Retrieval, RAG & Knowledge Systems
Building knowledge-augmented AI systems.
| Topic | Description |
|---|---|
| Vector Databases | Semantic search and embeddings |
| Full Text Search Deep Dive | Search engine internals |
| Text Search Tools | Search implementation |
| Apache Lucene Deep Dive | Lucene search library |
7. LLM Application Engineering
Building applications with LLMs.
| Topic | Description |
|---|---|
| Pydantic AI | Structured outputs with LLMs |
| AI SDK Projects | Building with AI SDKs |
| Prompt Engineering | Effective prompting techniques |
| LangChain Projects | LangChain framework |
| AI Agents | Building autonomous agents |
8. LLM Systems, Performance & Infrastructure
Scaling and optimizing AI systems.
| Topic | Description |
|---|---|
| AI Systems Deep Dive | Production AI infrastructure |
| Apache Arrow Deep Dive | Columnar data format |
| Apache Parquet Deep Dive | Efficient data storage |
| Apache Spark Deep Dive | Distributed data processing |
| Realtime Analytics Databases | OLAP systems |
9. Evaluation, Reliability & Production AI
Making AI systems reliable in production.
| Topic | Description |
|---|---|
| Observability & Reliability | Monitoring AI systems |
| Performance Engineering | Optimizing performance |
| Threat Detection & Logging | Security and auditing |
10. Advanced / Research-Level Mastery
Cutting-edge AI research and development.
| Topic | Description |
|---|---|
| AI Research Deep Dive | Reading and implementing papers |
| AI SDK Projects (Advanced) | Advanced SDK patterns |
| AI Agents (Advanced) | Complex agent architectures |
Learning Tips
- Math first: Don’t skip the foundations - they pay dividends forever
- Code everything: Implement algorithms from scratch before using libraries
- Read papers: ArXiv is your friend - read the original research
- Build projects: Theory without practice is useless
- Stay current: This field moves fast - follow key researchers