# Vasilis Katsoulis

> Software engineering leadership, architecture, artificial intelligence, modern development practices, and lifelong learning.

Canonical site: https://vasili.protepo.com/
Sitemap: https://vasili.protepo.com/sitemap.xml

## Primary pages

- [Home](https://vasili.protepo.com/): Professional identity and current focus.
- [About](https://vasili.protepo.com/about/): Experience, leadership approach, expertise, and career impact.
- [Site Architecture](https://vasili.protepo.com/praxis/site-architecture/): Site architecture, implementation choices, diagrams, and root project structure.
- [Noesis](https://vasili.protepo.com/noesis/): Position papers, essays, engineering notes, and practical guidance.
- [Praxis](https://vasili.protepo.com/praxis/): Working software and engineering artifacts.
- [Contact](https://vasili.protepo.com/contact/): Official contact options and verified professional profiles.

## Noesis

- [The Dawn of AI-Augmented Engineering](https://vasili.protepo.com/content/7/dawn-of-ai-augmented-engineering/): Software engineering is entering a new period in which artificial intelligence participates directly in analysis, implementation, testing, diagnosis, documentation, and learning.
- [From AI Demos to AI Capability](https://vasili.protepo.com/content/4/from-ai-demos-to-ai-capability/): How teams can move from impressive prototypes to maintainable AI-enabled systems.
- [Why Build a Tiny Transparent AI System?](https://vasili.protepo.com/content/11/why-build-a-tiny-transparent-ai-system/): Why small, inspectable AI systems help engineers understand tokenization, embeddings, tensors, runtime behavior, and responsible AI architecture.
- [How an AI System Sees a Sentence](https://vasili.protepo.com/content/15/how-an-ai-system-sees-a-sentence/): How tokenization turns a business question into discrete text units, why token boundaries matter, and what MAGPAI makes inspectable before token IDs and embeddings enter the story.
- [The AI and Machine-Learning Stack](https://vasili.protepo.com/content/13/the-ai-and-machine-learning-stack/): How AI-enabled applications span business workflows, orchestration, learned model behavior, numerical runtimes, infrastructure, and mathematical foundations.
- [The Engineering Leader in the Age of AI](https://vasili.protepo.com/content/6/engineering-leader-age-of-ai/): AI raises the standard for architecture, judgment, and organizational clarity.
- [Personal Knowledge Systems for Software Engineers](https://vasili.protepo.com/content/5/personal-knowledge-systems-software-engineers/): A practical case for treating notes, diagrams, and decision records as engineering infrastructure.

## Praxis

- [Introducing MAGPAI: A Tiny Transparent AI System](https://vasili.protepo.com/content/12/introducing-magpai-a-tiny-transparent-ai-system/): A transparent educational AI laboratory that demonstrates how language becomes tokens, vectors, tensors, runtime traces, and data-backed results.
- [Learning Clock](https://vasili.protepo.com/content/9/learning-clock/): A desktop learning telemetry tool integrated with a Markdown knowledge vault.
- [MAGPAI Stack Explorer](https://vasili.protepo.com/content/14/magpai-stack-explorer/): An accessible visual guide to the sixteen layers beneath a MAGPAI-style AI-enabled application, from business intent to models, compute, mathematics, operations, and governance.
- [MAGPAI Tokenizer Lab](https://vasili.protepo.com/content/16/magpai-tokenizer-lab/): An inspectable Tokenizer Lab that shows how the recurring MAG business question becomes normalized text, tokens, token IDs, and vectors in the MAGPAI Session 01 teaching pipeline.

## Identity

- [LinkedIn](https://www.linkedin.com/in/vasilis)
- [GitHub](https://github.com/javaboy-vk)