AI and machine learning topics
Browse every lesson by subject. Each topic is free, opens without a login, and combines explanation, an interactive visual, hands-on practice, and deeper references.
Guided topic hubs
Follow a coherent route when you want more than a single lesson.
Beginner learning hub
Machine Learning Foundations
Learn machine learning from problem framing and data preparation through baselines, evaluation, linear models, and responsible deployment decisions.
7 ordered lessonsLanguage model learning hub
Transformers and Large Language Models
Trace how text becomes tokens, contextual representations, next-token probabilities, decoded outputs, and evaluated language-model behavior.
7 ordered lessonsGrounded AI learning hub
RAG and Knowledge Systems
Learn embeddings, hybrid retrieval, RAG evaluation, ontologies, knowledge graphs, GraphRAG, and evidence-grounded agent architecture.
6 ordered lessonsReliable agents learning hub
AI Agents, Reliability, and Safety
Build bounded AI agents with typed tools, explicit state, MCP interfaces, harnesses, evaluation, security boundaries, and recoverable execution.
7 ordered lessonsSystems learning hub
Production AI Systems
Learn the training, memory, inference, optimization, serving, and observability trade-offs behind production AI workloads.
7 ordered lessonsAI & ML Foundations
9 topics- AI & ML OrientationWhat AI and ML are, how learning differs from rules, and how to choose a first path.
- Python & NumPy for MLWrite small Python programs and reason about arrays, shapes, vectorization, and reproducible experiments.
- Data Exploration & PreprocessingInspect real data, handle missing values, encode features, avoid leakage, and build repeatable transformations.
- Calculus & GradientsUnderstand slopes, partial derivatives, the chain rule, and gradient descent through visual and coded experiments.
- ML Workflow & EvaluationFrame a problem, split data correctly, choose metrics, detect overfitting, and compare against a baseline.
- Responsible AI FoundationsTranslate fairness, privacy, safety, transparency, and human oversight into testable design decisions.
- Logistic RegressionBuild an interpretable binary classifier using logits, sigmoid probabilities, cross-entropy, and thresholds.
- Model Evaluation ProjectAudit two candidate classifiers and defend a deployment threshold using evidence rather than one headline score.
- Beginner ML CapstoneTake a small tabular problem from question and data audit through model card, test evidence, and deployment recommendation.
Mathematics & Statistics
5 topics- ProbabilityUnderstanding randomness and Bayesian vs Frequentist views.
- DistributionsNormal, Binomial, and Poisson distributions.
- Linear AlgebraVectors, Matrices, Eigenvalues, and SVD.
- Hypothesis Testingp-values, Z-tests, and T-tests.
- Convex Optimization & KKTLagrange multipliers, KKT optimality conditions, Slater condition, and duality.
Classical Machine Learning
7 topics- Linear RegressionPredicting continuous values using a linear relationship.
- Decision TreesHierarchical split-based classification and regression.
- Tree EnsemblesCombine trees with bagging and boosting, evaluate out-of-sample behavior, and diagnose feature importance safely.
- SVMSupport Vector Machines for optimal boundary separation.
- K-Means ClusteringUnsupervised grouping of data points.
- Statistical Learning TheoryVC Dimension, Rademacher complexity, generalization error bounds, and PAC learning.
- Graph Neural Networks (GNNs)Message passing paradigms, Graph Convolutional Networks (GCN), GraphSAGE, and non-Euclidean graph representations.
Deep Learning
6 topics- Neural NetworksThe building blocks of deep learning: Perceptrons and MLPs.
- BackpropagationHow neural networks learn using the chain rule.
- RNNs & LSTMsBackpropagation Through Time (BPTT), vanishing gradients, forget gates, and long-range sequence memory.
- CNNsConvolutional Neural Networks for image recognition.
- TransformersThe attention mechanism that powers modern AI.
- Attention Variants & NormsGrouped-Query Attention (GQA), Multi-Query (MQA), RoPE embeddings, and RMSNorm topology.
Generative & Agentic AI
17 topics- How LLMs Generate TextSee how a language model predicts, selects, and appends one token at a time.
- LLM TokenizationUnderstand how text becomes versioned token IDs, context usage, and model cost.
- Embeddings & Context WindowsLearn how models represent text and how finite context changes application design.
- LLM Decoding & SamplingCompare greedy decoding, temperature, and top-p through visible probability changes.
- Prompting & In-Context LearningWrite prompts as testable contracts with instructions, examples, data, and output rules.
- LLM Limitations & GroundingRecognize unsupported output and design evidence, citations, abstention, and evaluation.
- LLMsLarge Language Models and the GPT architecture.
- RAGRetrieval Augmented Generation for grounding AI.
- RLHFReinforcement Learning from Human Feedback.
- Frontier Reasoning & GRPODeepSeek-R1 style Group Relative Policy Optimization, rule-based RL, and test-time compute.
- Agentic AIReasoning loops, tools, and multi-agent systems.
- Generative ModelsGANs, VAEs, and Diffusion Probabilistic Models (DDPM/DDIM).
- Flow Matching & DiT VideoRectified Flow matching, optimal transport vector fields, and Patchified Diffusion Transformers (Sora/Flux).
- State Space Models & MambaContinuous-time SSMs, selective scan parameters, discretization, and linear-time sequence modeling.
- Multimodal AI & CLIPContrastive vision-text embeddings, InfoNCE loss, and joint cross-modal representation spaces.
- Mixture of Experts (MoE)Sparse routing mechanisms, gating networks, Top-K dispatch, and auxiliary load balancing.
- Mechanistic InterpretabilitySparse Autoencoders (SAEs), monosemantic feature dictionaries, circuit analysis, and activation steering.
RAG & Knowledge Systems
5 topics- Embeddings & Hybrid SearchTokenization, dense and sparse representations, similarity, ANN indexes, quantization, and rank fusion.
- LLM Interfaces & Structured OutputsContext construction, prompt contracts, tool calls, schema validation, retries, and observable model APIs.
- Knowledge Graphs & OntologiesRDF triples, Turtle, OWL, SPARQL, SHACL, taxonomies, schemas, and semantic reasoning.
- Production RAG & GraphRAGChunking, query transformation, hybrid retrieval, reranking, entity resolution, and global graph search.
- Knowledge-Driven Agent SystemsOntology-constrained plans, graph-augmented reasoning, live knowledge updates, and enterprise architecture.
Agent Engineering & Safety
6 topics- Model Context Protocol (MCP)Clients, servers, tools, resources, prompts, transports, capability negotiation, authorization, and safe execution.
- Loop EngineeringBounded state machines, termination contracts, recovery, idempotency, budgets, no-progress detection, and replayable agent loops.
- Agent State, Memory & OrchestrationState graphs, checkpoints, working and semantic memory, durable workflows, interrupts, and multi-agent routing.
- Harness Engineering & SandboxesReproducible execution, containers and microVMs, resource limits, mocks, structured observations, and trace capture.
- AI Evaluation & Quality GatesGolden datasets, RAG and agent metrics, offline and online evaluation, regression tests, and CI release gates.
- Agent Guardrails & SecuritySchema controls, prompt-injection defense, least privilege, circuit breakers, budgets, approvals, and incident response.
Production AI
4 topics- Distributed TrainingParallelism paradigms (DDP, TP, PP, FSDP) and scaling laws.
- Model ServingHigh-throughput LLM hosting, KV caching, vLLM, speculative decoding, and quantization.
- Quantization & LoRALow-Rank Adaptation (ΔW = B·A), rank selection, FP8/INT4 uniform quantization, and scale-zero point math.
- FlashAttention & GPU ArchitectureGPU HBM vs SRAM bandwidth, online softmax tiling, IO-awareness, and Triton CUDA ops.