Understand how Transformer models predict the next token using embeddings, causal self-attention, KV caching, and autoregressive decoding.
Category: software engineering
LLM Inference Engine (chapter 5): Transformer Architecture From Scratch – Feed Forward Sub-Layer
Learn how the Transformer FFN works, including RMSNorm, Gate and Up projections, SiLU activation, element-wise multiplication, Down projection, and residual connections.
LLM Inference Engine (chapter 4): Transformer Architecture From Scratch – Attention Scores
Learn how to calculate Transformer attention scores step by step.
LLM Inference Engine (chapter 3): Transformer Architecture From Scratch – Token Embedding
Learn how the Transformer embedding layer converts token IDs into numerical vectors, builds embedding matrices, and prepares text for LLM processing using Llama 3 as an example.
LLM Inference Engine (chapter 2): Transformer Architecture From Scratch – Tokenization
Learn about Decoder-only Transformer Architecture, foundational matrix operations, and Tokenization step with llama3 model
LLM Inference Engine (chapter 1): Transformer Architecture From Scratch – GGUF Format
Introduction to Inference Engine and GGUF format (LLM model)
Build a Private RAG System with Local Llama
Learn how to build a private RAG (Retrieval-Augmented Generation) system using a locally deployed Llama model. This step-by-step guide covers document processing, embeddings, vector databases, semantic search, and prompt construction to help an AI assistant answer questions using sensitive internal documents without sending data to external AI services.
What is WebView and Compliance Concerns around WebView
What is a WebView, how do hybrid apps use it, and which PCI DSS, GDPR and HIPAA risks does it create? Plus a checklist to reduce them in mobile apps.
What Makes Social Networks Addictive? And What We Can Learn For Software Development.
Infinite scroll, variable rewards and notifications are built to keep you scrolling. How social networks hook users, and what developers can learn from it.
9 Reasons Why You Never Hit Your Goals (And What Actually Works)
Why do most goals fail? Nine reasons behind abandoned goals, from vague targets to broken feedback loops, and the habits that actually make progress stick
