Notes on .NET, XAF, and building things
Twenty years of XPO, application frameworks, and lately a lot of applied AI. 358 articles and counting.
Browse by topic
All topics →Latest articles
View all →
Agents That Defend Goals: What Two Scientists and an Engineer Taught Me About Building Agents
I turned three accounts of what a mind is, and who should own one, into agents in C#: a colony of controllers from Michael Levin, a self-model from Joscha Bach, a hybrid of both, and an owned critic from George Hotz. Then I put them in worlds that lie and on real code with a live model in the seat. The biology turned out to be scaffolding. Five rules did the work, and one of them carried all the weight: write the goal as evidence the agent cannot author.
Agents That Defend Goals, Part 7: The citizen's agent
George Hotz's rules for owning a machine, turned into a C# critic that earns the right to interrupt, cites the registry, and answers to one person.
Agents That Defend Goals, Part 6: Worlds that lie
How the permit office, the irrigation turn and the cattle lot are built so that the registry, not the agent, is the judge. And how to build the next one.
Agents That Defend Goals, Part 5: The control layer on Microsoft Agent Framework
Five rules as function middleware around a real model, and the benchmark finding that the plumbing bounds the architecture.
Agents That Defend Goals, Part 4: The careless agent, and why the goal must be evidence
The comparison agent every demo runs against, why it fails the way real agents fail, and the one finding from the benchmark that carries all the weight.
More topics
All topics →
Agents That Defend Goals, Part 3: One error channel at two altitudes
Levin's body with Bach's mind: the hybrid agent, and the single line it rests on.
Agents That Defend Goals, Part 2: The agent that reads itself first
Joscha Bach's account of the mind, turned into a C# agent that keeps a falsifiable record of itself and earns the right to act.
Agents That Defend Goals, Part 1: An agent as a colony of controllers
Michael Levin's biology, turned into a C# agent, and what it does when the office lies to it.
What Are Embeddings? King − Man + Woman, Tested for Real
An embedding turns a word or a sentence into a list of numbers where similar meanings land close together. The famous party trick is king − man + woman = queen. I tested it on real vectors: classic GloVe gets queen, and so does a modern LLM's embedding table, but each fails somewhere interesting. Then the useful part: a sentence about rain in English, Spanish, Russian and Chinese lands in almost the same place, even though the tokenizer cut each one into completely different pieces.
What Is Attention? How a Word Decides Which Other Words Matter
Attention is how a language model lets each word look back at the others and decide which ones matter. I read the real attention weights out of a small open model for 'The animal didn't cross the street because it was too tired' and the same sentence ending in 'wide'. In one head, 'tired' puts 73% of its attention on 'animal'. Change one word, and 34 of the 336 heads switch toward 'street'. Plus the surprise: 61% of all attention goes to the very first word.