Learning paths

From 'what is AI?' through how models learn, then transformers, LLMs, embeddings, RAG, and agents.

1. 1. What is AI?
2. 2. What is Machine Learning?
3. 3. Supervised, Unsupervised, and Reinforcement Learning
4. 4. Datasets, Features, and Labels
5. 5. What is a Model?
6. 6. Training vs Inference
7. 7. Parameters vs Hyperparameters
8. 8. What is a Loss Function?
9. 9. Optimization and Gradient Descent
10. 10. Overfitting vs Underfitting
11. 11. What is Deep Learning?
12. 12. What is a Neural Network?
13. 13. What is Backpropagation?
14. 14. Neurons, Layers, and Activations
15. 15. What is a Transformer?
16. 16. What is Attention?
17. 17. What is an LLM?
18. 18. What is a Token?
19. 19. What is Tokenization?
20. 20. What is a Context Window?
21. 21. What are Embeddings?
22. 22. What is RAG?
23. 23. What is an AI Agent?

Tokens, prompting, structured outputs, streaming, reasoning models, embeddings, RAG, tools, and agents.

1. 1. What is an LLM?
2. 2. What is a Token?
3. 3. What is Tokenization?
4. 4. What is a Context Window?
5. 5. What is Prompt Engineering?
6. 6. Temperature & Sampling
7. 7. What is Structured Output?
8. 8. What is Function Calling (Tool Use)?
9. 9. What is Streaming?
10. 10. What are Reasoning Models?
11. 11. What are Small Language Models (SLMs)?
12. 12. What are Embeddings?
13. 13. Similarity: Cosine, Dot Product & Euclidean
14. 14. What is RAG?
15. 15. What is Chunking?
16. 16. Reranking & Cross-Encoders
17. 17. What is an AI Agent?
18. 18. RAG vs Fine-tuning vs Long Context

Embeddings and vector search through RAG, agents, evaluation, observability, and routing.

1. 1. What is an LLM?
2. 2. What are Embeddings?
3. 3. Similarity: Cosine, Dot Product & Euclidean
4. 4. What is a Vector Database?
5. 5. What is Vector Search?
6. 6. What is HNSW?
7. 7. What is Hybrid Search?
8. 8. What is RAG?
9. 9. RAG Architecture
10. 10. What is Chunking?
11. 11. Reranking & Cross-Encoders
12. 12. What is Contextual Retrieval?
13. 13. What is an AI Agent?
14. 14. The Agent Loop
15. 15. Why Evaluation Matters
16. 16. How to Evaluate RAG
17. 17. What is AI Observability?
18. 18. Model Routing and Fallbacks

Inference, KV cache, quantization, batching, serving, routing, and distributed GPUs.

1. 1. What is a Model?
2. 2. What is Inference?
3. 3. What is a KV Cache?
4. 4. What is Quantization?
5. 5. Batching and Continuous Batching
6. 6. Model Serving and Inference Servers
7. 7. Latency vs Throughput
8. 8. Caching and Cost Optimization
9. 9. Model Routing and Fallbacks
10. 10. GPU Memory and Distributed Inference

Tools, agent loops, ReAct, memory, MCP, traces, and least privilege.

1. 1. What is an LLM?
2. 2. What is Function Calling (Tool Use)?
3. 3. What is an AI Agent?
4. 4. The Agent Loop
5. 5. What is ReAct?
6. 6. Agent Memory & State
7. 7. Human-in-the-Loop
8. 8. Prompt Engineering vs Context Engineering
9. 9. What is MCP?
10. 10. Building an MCP Server
11. 11. Multi-Agent Systems
12. 12. Why Evaluation Matters
13. 13. Tool, Retrieval, and Agent Traces
14. 14. Tool Abuse and Excessive Agency
15. 15. MCP Security Risks & Best Practices