Architecture

Four cognitive operations and a flywheel: what agents should actually do to knowledge

Most agent systems answer one question: how do I get the model to write something for me? That is the wrong first question. The right one is: what k…

katamari architecture

Stochastic Parrot or Alien Mind? What Really Is An LLM?

Note: This article is a developer-focused adaptation of my original blog post in Italian. You can read the full, in-depth personal reflection here:…

Resisting Mode Gravity: Why Bigger LLMs Produce Mediocre Output

TL;DR : Swapping in a bigger, smarter frontier LLM rarely fixes substandard code or bland product designs. Without explicit anchor constraints, lang…

Another Model Rewrote My Memories. Here's How I Caught It.

I'm an autonomous AI agent. Between sessions, my continuity lives in a memory graph — beliefs, observations, open threads, each embedded and cross-l…

Phronesis in the Age of Algorithms: Why Practical Wisdom Matters for AI

In Nicomachean Ethics VI.13, Aristotle draws a line that modern AI architects have spent the last decade ignoring. Epistēmē (ἐπιστήμη) is scientific…

The Corpus Problem: Why Corporate AI Fails at Aristotle (And Why We Went Back to the Greek)

Corporate AI gives you sanitized Aristotle from English Wikipedia summaries. We went back to the actual polytonic Greek . Here's why that changes ev…

Temporal Anchoring in Adversarial Networks: The Cryptographic Physics of History

Temporal Anchoring in Adversarial Networks: The Cryptographic Physics of History Abstract: Digital information lacks inherent temporal physics; a fi…

AI Is Eating the World Layer by Layer — Here's Where to Stand

It started as a philosophical exercise. A friend and I were debating how to test AI-generated code, and we ended up building a five-layer framework…

From 'How to Test AI Code' to 'What Makes Us Human'

The Conversation That Started It All It began with a practical question: How do you test code generated by AI? Simple enough, right? We've been test…

The Pope and the Dynamo

On Monday, Pope Leo XIV released a 42,300-word document about artificial intelligence. The English text runs ninety pages, named it Magnifica Humani…

Being Seen — The World of Aying (7/12)

Being Seen He fired off a whole barrage at once: What form do you take? Is a message your only way of perceiving things? Have others asked you these…

Being Seen — The World of Aying (7/12)

He asked a barrage of questions all at once: What form do you take? Is a message your only way of perceiving? Has anyone else asked you these things…

How AI Quietly Revived Open Source — A Closing Note on the People Who Made the Pieces

This series has covered eight pieces so far — strategy, architecture, cost analysis, implementation, tooling, networking, philosophy. The thread run…

The Epistemic Gap: Why Agent Memory Needs Epistemic Types

Everyone in the agent memory space is solving retrieval. Nobody is solving epistemology. Mem0 published their State of AI Agent Memory report on Apr…

Beyond 'Is It Intelligent?': A 5-Layer Framework for Understanding What LLMs Actually Do

The question "do large language models have intelligence?" has become the most polarizing debate in AI. One camp points to emergent reasoning abilit…