Chuyển đến nội dung chính
BlogCoursesHL7Cert PrepRoadmapGameDomainAboutMCPNEW
Topics
DatabaseDevOpsKiến trúc hệ thốngLinuxProgrammingSecurity

AI & LLM: From Basics to Advanced

Comprehensive course on Artificial Intelligence and Large Language Models — from Neural Networks platform, Transformer architecture, to Fine-tuning, RAG, AI Agents and production deployment. Learn hands-on with Python, PyTorch, and LLM APIs.

Introducing the Series

The AI & LLM: From Basics to Advanced series is a comprehensive journey from the most fundamental concepts of AI to building and deploying production-ready LLM applications.

You will not only learn theory but also practice code of each concept — from writing Neural Networks with NumPy, implementing Attention Mechanism with PyTorch, fine-tuning LLM on GPU, building RAG pipelines, to deploying models to real servers.

What will you learn?

Part 1: AI & Deep Learning Platform

  • Lesson 1: Overview of AI, ML, DL and LLM — the big picture
  • Lesson 2: Mathematics for AI: Linear Algebra, Calculus, Probability
  • Lesson 3: Neural Networks from scratch: Perceptron, Backpropagation, Activation Functions
  • Lesson 4: Deep Learning: CNN, RNN, LSTM — the premise of Transformer

Part 2: Transformer architecture

  • Lesson 5: Attention Mechanism: Self-attention and Multi-head Attention
  • Lesson 6: Transformer architecture: Encoder, Decoder, Positional Encoding
  • Lesson 7: BERT and Encoder-only Models: Masked Language Modeling
  • Lesson 8: GPT and Decoder-only Models: Autoregressive generation
  • Lesson 9: Tokenization: BPE, WordPiece, SentencePiece

Part 3: Training & Fine-tuning LLMs

  • Lesson 10: Pre-training: CLM, MLM and Scaling Laws
  • Lesson 11: Supervised Fine-Tuning (SFT): Instruction Tuning
  • Lesson 12: PEFT: LoRA, QLoRA — efficient fine-tuning with few resources
  • Lesson 13: RLHF and Alignment: DPO, PPO, Constitutional AI

Part 4: Prompting & RAG

  • Lesson 14: Prompt Engineering: Zero-shot, Few-shot, System Prompts
  • Lesson 15: Advanced Prompting: Chain-of-Thought, Tree-of-Thought, ReAct
  • Lesson 16: RAG: Retrieval Augmented Generation from A to Z
  • Lesson 17: Vector Databases: Embeddings, Semantic Search, ChromaDB, Qdrant

Part 5: Building AI Applications

  • Lesson 18: AI Agents: Tool Use, Function Calling, Agentic Workflows
  • Lesson 19: LLM APIs: OpenAI, Anthropic Claude, Google Gemini

Part 6: Production & Enhancement

  • Lesson 20: Deploying LLMs: Ollama, vLLM, TGI & Evaluation
  • Lesson 21: Running AI Local with Ollama on Apple Silicon — Deep Dive

Input required

  • Basic Python (know how to write functions, classes, list comprehension)
  • Level 3 math (no need to specialize — lesson 2 will review what is needed)
  • Computer with at least 8GB RAM (GPU not required for early parts)

Tools used

Python 3.11+     | Ngôn ngữ lập trình chính
NumPy / Pandas   | Xử lý số học và dữ liệu
PyTorch          | Deep Learning framework
Hugging Face     | Transformers, Datasets, TRL
LangChain        | LLM application framework
Ollama           | Chạy LLM local
OpenAI API       | GPT-4o, Embeddings
Anthropic API    | Claude
ChromaDB / FAISS | Vector databases

Part 1: AI & Deep Learning Platform

Lesson 1: Overview of AI, ML, DL and LLM90 minLesson 2: Mathematics for AI — Linear Algebra, Calculus, Probability120 minLesson 3: Basic Neural Networks — Perceptron & Backpropagation150 minLesson 4: Deep Learning Overview — CNN, RNN, LSTM120 min

Part 2: Transformer architecture

Lesson 5: Attention Mechanism — Self-attention & Multi-head Attention150 minLesson 6: Transformer architecture — Encoder, Decoder, Positional Encoding180 minLesson 7: BERT and Encoder-only Models150 minLesson 8: GPT and Decoder-only Models150 minLesson 9: Tokenization — BPE, WordPiece, SentencePiece90 min

Part 3: Training & Fine-tuning LLMs

Lesson 10: Pre-training LLMs — CLM, MLM and Scaling Laws120 minLesson 11: Supervised Fine-Tuning (SFT) — Instruction Tuning180 minLesson 12: PEFT — LoRA, QLoRA and Adapter Methods180 minLesson 13: RLHF and Alignment — DPO, PPO150 min

Part 4: Prompting & RAG

Lesson 14: Prompt Engineering — Zero-shot, Few-shot, System Prompts120 minLesson 15: Advanced Prompting — CoT, ToT, ReAct120 minLesson 16: RAG — Retrieval Augmented Generation from A to Z210 minLesson 17: Vector Databases — Embeddings & Semantic Search150 min

Part 5: Building AI Applications

Lesson 18: AI Agents — Tool Use, Function Calling, Agentic Workflows210 minLesson 19: LLM APIs — OpenAI, Anthropic Claude, Google Gemini150 min

Part 6: Production & Enhancement

Lesson 20: Deploying LLMs — Ollama, vLLM, TGI & Evaluation210 minLesson 21: Running AI Local with Ollama on Apple Silicon — Deep Dive180 min