Learn AI. Know it by heart.

Understand modern AI concepts through simple explanations, interactive visuals, real examples, and hands-on learning — in one calm, reading-first place.

What is RAG?What is an embedding?How does attention work?What is MCP?

Explore AI

FoundationsThe base ideas behind all of AI: models, training, and inference.LLMsLarge language models: tokens, context, prompting, and tool use.EmbeddingsTurning meaning into vectors you can compare mathematically.Vector DatabasesStoring and searching embeddings at scale.RAGRetrieval-Augmented Generation: grounding models in real information.AgentsAI systems that reason, plan, and use tools to reach goals.MCPThe Model Context Protocol connecting models to tools and data.AI System DesignDesigning complete, production-grade AI systems.

Browse the full curriculum

Choose your path

AI BeginnerFrom 'what is AI?' through how models learn, then transformers, LLMs, embeddings, RAG, and agents.23 lessons · ~12h LLM DeveloperTokens, prompting, structured outputs, streaming, reasoning models, embeddings, RAG, tools, and agents.18 lessons · ~14h AI EngineerEmbeddings and vector search through RAG, agents, evaluation, observability, and routing.18 lessons · ~18h AI Systems EngineerInference, KV cache, quantization, batching, serving, routing, and distributed GPUs.10 lessons · ~12h AI Agent EngineerTools, agent loops, ReAct, memory, MCP, traces, and least privilege.15 lessons · ~16h

Start reading

The Agent Loop Agent Memory & State Agent vs Chatbot vs Workflow Human-in-the-Loop Multi-Agent Systems Planning & Reflection What is ReAct?Tool Selection What is an AI Agent?Design an AI Search Engine Design a Coding Agent Design an Enterprise Knowledge Assistant Design a RAG Chatbot Design a Research or Multi-Agent System Design a Customer-Support Agent How to Design an AI System Context Caching Context Compression Selection & Prioritization State Management Long-Context Strategies Prompt Engineering vs Context Engineering What are Autoencoders?What are CNNs?What are GANs?LSTMs and GRUs Neurons, Layers, and Activations What are RNNs?Why Transformers Replaced RNNs Dense vs Sparse Embeddings Embedding Models & Dimensions What are Matryoshka Embeddings?Similarity: Cosine, Dot Product & Euclidean What are Embeddings?Benchmarks, Gold Sets, and Regression Tests Hallucination and Safety Evaluation Human Evaluation LLM-as-a-Judge Offline vs Online Evaluation Quality, Latency, Cost, and Reliability Why Evaluation Matters Full Fine-tuning vs PEFT Instruction Tuning vs Preference Tuning LoRA and QLoRA What is Quantization?What is Fine-tuning?When to Fine-tune vs RAG vs Prompting What is Backpropagation?Datasets, Features, and Labels What is a Loss Function?Optimization and Gradient Descent Parameters vs Hyperparameters Training vs Inference What is AI?What is Deep Learning?What is Machine Learning?What is a Model?What is a Neural Network?Batching and Continuous Batching GPU Memory and Distributed Inference Caching and Cost Optimization KV Cache at Serving Time Latency vs Throughput Model Routing and Fallbacks Model Serving and Inference Servers Quantization for Serving What is Inference?What is a Context Window?What is Function Calling (Tool Use)?What is Prompt Engineering?What are Reasoning Models?What are Small Language Models (SLMs)?What is Streaming?What is Structured Output?Temperature & Sampling What is Tokenization?What is an LLM?What is a Token?Precision, Recall, F1, and ROC-AUC What is Classification?What is Clustering?What is Cross-Validation?What are Decision Trees?Gradient Boosting and XGBoost Overfitting vs Underfitting What are Random Forests?What is Regression?Supervised, Unsupervised, and Reinforcement Learning Building an MCP Server Consuming an MCP Server MCP + Agents, IDEs, and Products MCP Architecture MCP Authentication & Authorization Tools, Resources, and Prompts MCP Security Risks & Best Practices What is MCP?Why MCP Exists Long-term Memory Memory Conflicts & Security Storage, Retrieval & Forgetting Semantic, Episodic & Procedural Memory Short-term Memory & Conversation History User Memory vs Agent Memory Audio & Video Understanding Image Understanding & OCR Image & Video Generation Multimodal Agents Multimodal Embeddings & RAG Speech-to-Text & Text-to-Speech Vision-Language Models Tool, Retrieval, and Agent Traces Prompt Versioning and Evaluation Traces Token Usage, Latency, and Cost Tracking Tracing, Logging, and Metrics What is AI Observability?Agentic RAG & Advanced Patterns What is Chunking?What is Contextual Retrieval?What is Query Transformation?RAG Architecture How to Evaluate RAG RAG vs Fine-tuning vs Long Context Reranking & Cross-Encoders What is RAG?Data Leakage and Sensitive Disclosure Indirect Prompt Injection Insecure Output Handling Jailbreaks Agent Permission Boundaries What is Prompt Injection?RAG Poisoning Secrets Management and Sandboxing Tool Abuse and Excessive Agency What is a KV Cache?What is Multi-Head Attention?What is Positional Encoding?What is Self-Attention?What is Attention?What is a Transformer?Exact vs Approximate Nearest Neighbor What is HNSW?What is Hybrid Search?IVF & Product Quantization What is Metadata Filtering?What is Vector Search?What is a Vector Database?

Every lesson ends with one essential idea to remember — so you don't just read it, you know it by heart.