AI Engineering and Architecture

Engineering notes on building AI systems whose behavior can be understood, evaluated, and improved systematically.

Notes by Valery Yuda. Email: valery.yuda@gmail.com. GitHub: valery-judah.

Content access

Some notes on this site may be intentionally obfuscated to discourage unauthorized automated scraping and use in LLM training. This is a temporary measure while I look for a more appropriate way to control machine access to and reuse of this content.

Selected Readings

A small selection of notes and drafts to start with.

AI Systems Require Behavioral Evidence

Why the quality of an AI system must be observed and measured under realistic conditions.

Evaluation

Drafts on evaluating AI systems, understanding their failures, and improving their behavior.

Natural Language Interface Substrate

A working model of how meaning, context, ambiguity, and intent shape natural-language interfaces.

Browse by Section

Use the site by major topic area when you want the broader map instead of a single highlighted note.

AI Engineering

Practical notes on LLM-based systems, evaluation workflows, and agent behavior.

System Design

Architecture, trade-offs, distributed systems, and interview-oriented design frameworks.

Engineering

General software engineering practices, architecture thinking, and implementation discipline.

Projects

Worked engineering concepts and design explorations grounded in practical system concerns.

Start with AI Engineering