The scaffolding is the product
Anthropic accidentally published 512,000 lines of Claude Code's source. The interesting part isn't the model — it's the context management plumbing around it.
All Posts
All
AI Engineering
LLM Patterns
Career & Thinking
3D Printing & Making
Infrastructure
Tools & Workflow
Context is the new automation layer
Business automation used to stall where the easy stuff ended and the judgment calls began. That's changing — not because AI replaces workflow tools, but because the people who unde...
The context wars: how Gemini, ChatGPT, Claude and Grok manage agent memory in 2026
I've spent months building context management patterns by hand. Here's how the four major platforms are solving the same problems at the infrastructure level — and what it means fo...
My .md files vs Claude's memory tool: a practitioner comparison
Anthropic shipped a memory tool that looks structurally identical to the markdown files I've been maintaining by hand. Here's what each approach gets right, where they diverge, and...
Custodians of context: how our roles must evolve in the AI era
I've been writing about context management at the individual and team level. This post zooms out: if context is the real bottleneck in AI, who owns it across an organization — and...
Separation of concerns in AI agent teams: lessons from multi-disciplinary delivery (and where they break)
We've been solving the 'too complex for one person' problem for decades in program delivery. Now some team members are language models — and the coordination overhead is measured i...
The recipe app I couldn't build six years ago
Six years ago I helped a CPG company dream up a smart meal planner. We never shipped it. This year, my wife and I built it — and we use it every week.
More agents, less cost
Running three AI agents sounds expensive. It's cheaper than running one — if you put the right model in each role.
When one context window isn't enough
Splitting a single AI agent into a team didn't just change the workflow — it changed how context flows through the whole system.
I was the validator all along
My AI coding setup worked — until I noticed I was spending more time checking output than building. The fix was giving the validation job to another agent.