All Articles

Deep Learning

Building Attention from Scratch

Implement the attention mechanism that powers transformers, from matrix operations to multi-head attention, with clear PyTorch code.

12 min read Advanced

LLMs

GPT from Scratch in 500 Lines

Build a miniature GPT model that actually generates coherent text. Covers tokenization, positional encoding, and training loops.

18 min read Advanced

Fundamentals

Vector Embeddings Explained

Why do LLMs represent words as vectors? Learn the intuition behind embeddings and build your own word2vec implementation.

10 min read Intermediate

Fundamentals

Backpropagation: The Math Behind Learning

Derive backpropagation from calculus first principles. Build a neural network training engine with automatic differentiation.

15 min read Advanced

Production

Serving LLMs at Scale

Deploy language models in production: quantization, batching, caching strategies, and monitoring for real-world applications.

14 min read Intermediate

Deep Learning

Convolutional Networks from First Principles

Build CNNs without frameworks. Understand convolutions, pooling, and feature maps by implementing them with NumPy.

16 min read Advanced

LLMs

Tokenization: From Text to Tokens

How do LLMs break text into pieces? Build BPE tokenization from scratch and understand subword algorithms.

9 min read Intermediate

Production

Fine-Tuning LLMs on Custom Data

Adapt foundation models to your domain. LoRA, prompt tuning, and full fine-tuning with practical examples.

13 min read Advanced

Fundamentals

Gradient Descent: Optimization Fundamentals

Visualize how neural networks learn. Implement SGD, Adam, and learning rate schedules from mathematical foundations.

11 min read Intermediate