Guide Articles

Step-by-step guides and tutorials for AI tools, frameworks, and implementations. Practical how-to content for developers and practitioners.

  • Home /
  • Guide Articles
Blueprint diagram showing three tokenizer library pathways converging into a unified vocabulary specification
MAX guide 12 min

How to Train and Choose a Custom Tokenizer with tiktoken, SentencePiece, and HF Tokenizers in 2026

Learn how to choose, train, and validate a custom tokenizer using tiktoken, SentencePiece, and HF Tokenizers with a …

Architectural blueprint of attention matrix computation showing QKV projection layers and optimization pathways
MAX guide 10 min

Implementing Attention from Scratch: PyTorch, FlashAttention, and Grouped-Query Optimization

Spec your attention implementation before writing code. Learn to decompose QKV projections, configure FlashAttention …

Engineer examining a vector search pipeline blueprint with index nodes and distance metric annotations on a diagnostic screen
MAX guide 11 min

Similarity Search Pipeline: FAISS, HNSWlib, ScaNN (2026)

Select between FAISS, HNSWlib, and ScaNN for production vector search. Specification-first approach covering index …

Architecture blueprints showing parallel encoder and decoder pathways with structured data flowing between them
MAX guide 11 min

When to Choose Encoder-Decoder Over Decoder-Only: T5, BART, and Whisper Use Cases in 2026

Learn when encoder-decoder models like T5, BART, and Whisper outperform decoder-only alternatives. A spec framework for …

Architectural blueprint of a transformer model with labeled attention layers and data flow arrows on a dark grid
MAX guide 12 min

How to Build a Transformer from Scratch Using PyTorch and Hugging Face

Specify a transformer from scratch in PyTorch and Hugging Face. Decompose attention, embeddings, and training loops into …

Specification blueprint overlaid with attention weight heatmaps flowing between token sequences
MAX guide 11 min

How to Implement Multi-Head Attention in PyTorch and Visualize Attention Patterns

Specify multi-head attention for AI-assisted PyTorch builds. Decompose QKV projections, constrain SDPA kernels, and …