Models, Context & Reasoning
How models behave inside real systems — architecture, context, training, and inference.
Neural Networks
Perceptrons, activations, backpropagation, and the path from simple neurons to deep architectures.
Under the HoodTokenization
BPE, SentencePiece, vocabulary tradeoffs, and how text becomes numbers.
Under the HoodTransformers
The transformer architecture, positional encodings, and scaling laws that power modern AI.
Under the HoodEmbeddings
Word2Vec, semantic similarity, RAG pipelines, and vector search fundamentals.
Under the HoodAttention
Query, Key, Value mechanics — Flash Attention, GQA, MQA, and multi-head patterns.
Under the HoodTraining Pipeline
Pre-training, RLHF, DPO, distributed training, and the full model development lifecycle.
Under the HoodInference Pipeline
Prefill/decode phases, KV-cache, speculative decoding, and serving optimization.
Under the HoodLoss & Gradient Descent
Loss functions, optimization algorithms, backpropagation, and regularization techniques.
Under the HoodProbability & Sampling
Softmax, temperature, beam search, top-k/top-p, and calibration methods.
Under the HoodData & Annotation
Data pipelines, RLHF labeling, active learning, and data governance for LLMs.
Under the HoodContext Engineering
17 prompt and context engineering techniques for building reliable AI systems.
Under the HoodHow LLMs Work — Complete Guide
End-to-end interactive visual overview of the entire LLM pipeline in one comprehensive page.
Under the Hood