Books

Memory From First Principles

Engineering AI Systems That Turn Past Experience Into Better Present Behaviour

Build AI memory from storage and retrieval through provenance, temporal state, unfinished work, context selection, behavioural evaluation, trust, and the boundary with learning — adding only the mechanisms that measurement earns.

Context From First Principles

Engineering What an AI Gets to Know

Build context systems from first principles: observe what a model actually receives, measure when more information hurts, then earn pruning, compaction, externalization, recall, caching, policy, budgeting, and a Context Compiler through controlled experiments.

Applied AI

Engineering Intelligence Into a Reliable Process

Build from a callable model to explicit context, durable state, controlled actions, independent verification, and a deterministic runtime that decides what happens next.

Embeddings From First Principles

Geometry, Retrieval, Alignment, and the Hidden Representation Layer of AI

Explore what it means for information to become geometry, when that geometry can be trusted, and how representation, similarity, retrieval, calibration, cross-space alignment, and compression combine into an embedding runtime that knows its own limits.

Browser AI From First Principles

Build browser-native AI from the capability boundary upward: local models, observability, WebMCP tools, agent security, and a user-owned browser policy engine.

DSPy From First Principles

Turn language-model behavior into an experimental variable, then learn when optimization is measurable, when improvement is trustworthy, and what DSPy contributes to the process.

Digital Life: From First Principles

An Experimental Search for Life-Like Organization in Computation

An experimental search for life-like organization in computational systems — through emergence, causality, persistence, material turnover, hidden state and finite computation.

Debugging AI

From Python Exceptions to Models, Hallucinations, and Agents

Progress from deterministic debugging to diagnosing models, evidence, trajectories, and AI-generated work.

Hallucination From First Principles

Explore why language models hallucinate, how hallucination can be measured and evaluated, where individual detection methods fail, and how evidence, verification, policy, abstention, and memory gates can be combined to build reliable systems around stochastic models.

PyTorch From First Principles

Understand PyTorch from first principles — from tensor geometry and autograd to model structure, transforms, attention, debugging, performance, compilation, reproducibility, and a GPT-style model built from scratch.

Models From First Principles

Learn to read modern AI models by decomposing them into smaller mechanisms, then rebuilding quality scorers, value and policy heads, recurrent reasoning systems, optimizers, and preference rankers in PyTorch.

Agents From First Principles

Build reliable AI agents by understanding the mechanisms underneath them: control loops, actions, validation, planning, state, tools, memory, search, and external verification.

Freestyle Cognition

Think, create, research, write, and build with AI through open conversation, clear intent, structured reflection, iterative collaboration, and amplified creation.

Agent Architectures: Advanced Strategies for Intelligent LLM Systems

Explore how LLM agents grow from simple conversations into reflective, tool-using, memory-enabled and multi-agent systems—and how those systems can become collaborators, companions, interfaces, and extensions of human agency.