Blog
Notes from the workbench.
In-depth, practical guides to building with AI — RAG architecture, choosing and evaluating LLMs, and connecting models to real systems with MCP. Written from what we've learned shipping them, without the hype.
Featured · Architecture
RAG Architecture, Explained: How to Build Retrieval-Augmented Generation That Actually Works
What RAG is, the two pipelines inside every RAG system, the design choices that decide answer quality — chunking, embeddings, hybrid search, reranking — and how to evaluate and run it in production.
Oct 4, 2026 · 14 min read
Oct 4, 2026
LLM Models, Benchmarks and Evals: How to Choose a Model on Evidence, Not Leaderboards
How large language models differ, what the popular benchmarks really measure — and where they mislead — and how to build your own evals so you pick and change models with confidence.
Models · 12 min read
Oct 4, 2026
MCP vs API: What the Model Context Protocol Is, Its Types, and How to Build and Deploy an MCP Server
MCP is the standard way for AI apps to use tools and data. How it differs from a normal API, its building blocks and server types, and a step-by-step guide to building an MCP server in Python and deploying it.
Protocols · 14 min read