Field Note
The Human in the loop is not a strategy
"Human oversight" is doing a lot of work in AI architecture diagrams. Too much, I think. It usually shows up as a reassuring little box near
Topic
AI
AI systems, engineering judgment, and the organisational choices behind useful automation.
Field Note
When the consumer is a machine
Event contracts used to be guidance. With agents on the other end, they're the thing the system actually runs on.
Field Note
Batch vs Streaming is the Wrong Data Architecture Question
The real architectural decision isn't how fast you process data. It's what you choose to treat as a fact, and whether the machines around it can reason about those facts without asking permission.
Field Note
Why the modern Data Platform is becoming an Operating System
Why it is no longer just infrastructure For a long time, the data platform was treated as a supporting layer. A place where data was collected, transformed, stored,
Field Note
AI development & Engineering clarity
A few years ago, most engineering organizations were constrained by the same thing: writing code was expensive. Not just in budget terms. Expensive in attention. In coordination. In
Field Note
Alfred does its homework before writing a line of code
There’s a particular kind of frustration engineers know well: you pick up a Jira ticket and the description reads something like “Fix the thing from the meeting.” No context. No requirements/details.
Field Note
I Let an AI Agent Read My Jira Tickets Before Reviewing PRs
How GitHub Agentic Workflows + MCP turned my pull request reviews from context-switching nightmares into fully-loaded briefings