You're signed in. Welcome to kevinsomany.com.
Grow your business, your career, and your opportunities with AI. I break real AI research into short reads and buildable projects, so you keep up with the technology without an engineering background.
Before an AI model is connected to any tool or document, it runs on pre-trained knowledge: patterns compressed from an enormous pile of public text. Here is what that background knowledge does well, where it fails, and how grounding fixes it.
Read article →Retrieval Augmented Generation
Why a trained model goes stale, how retrieval fixes it, and what teams actually build with RAG once they have it.
Agent Engineering
Seven builds that take an agent from a single loop to a coordinated, gated, resumable system. Each part is a working instrument, not a diagram.
Prompt instructions shape behaviour. Hooks decide what the application will allow. This project builds both hook points from scratch and shows why prevention has to sit before execution, not after it.
A better prompt can make a mistake rarer. A gate in your code can make a specific mistake impossible. This project shows where each belongs, and builds a refund tool that refuses to run until its prerequisites are met.
A good multi-agent system is not just the right subagents. It is a coordinator that can actually call them, and handoffs that carry every claim's origin with it. These are implementation problems, not theory problems.
Split a hard job across several AI agents and it can get faster or it can get chaotic. The difference is a coordinator that owns the plan, controls what each agent sees, and checks whether the combined result actually answers the request.
The gap between an AI that talks and an AI that does is a loop your program runs around the model. Here is that loop, built up one idea at a time and traced through a real support request, start to finish.
AI moves faster than anyone can follow alone. One short read at a time keeps you ahead of it.
For non-engineers eager to learn about AI engineering · Jargons explained · Based on the works of the most advanced AI contributors in the world · Free, unsubscribe anytime