Essays and technical notes on context memory, AI infrastructure, governed agents, and the systems layer emerging around intelligent software.
AI systems are becoming stateful. As applications and agents move beyond isolated prompts, context memory becomes a core production layer: something to inspect, correct, govern, and compose. A first note on what this layer should look like — and why we're building AtomicMemory openly.
A technical walkthrough of how AtomicMemory traces retrieval decisions, and why observability belongs in the data path — not next to it.
When the model was wrong about you yesterday and right today, where does that show up? Notes on versioned context.
Occasional, technical, low-frequency. No marketing. Unsubscribe at any time.