Articles
Thoughts on project management, AI and technology.
21 April 2026
10 MCP and Native Connections Every Knowledge Worker Should Try
Most MCP content is written for developers. Filesystem servers, GitHub, Sentry, Postgres. Useful if you write code. Not useful if you write proposals, run projects, or spend your week in Notion, Slack…
Read on Medium →
18 April 2026
Claude Design Is Here. Where Does It Fit in the Stack?
Anthropic launched Claude Design on April 17, 2026, in research preview for Pro, Max, Team, and Enterprise subscribers. It is the fourth core surface in Anthropic’s main app stack, after Chat, Cowork,…
Read on Medium →
15 April 2026
AI Code Ships Faster Than Teams Can Verify It
Zero percent. That is the share of senior SRE and DevOps leaders who described themselves as “very confident” that AI-generated code will behave correctly in production, according to Lightrun’s 2026 S…
Read on Medium →
14 April 2026
100 Robotaxis Froze at Once, and Not One Pulled Itself to Safety
On the evening of April 1, in the middle of rush hour in Wuhan, China, more than 100 of Baidu’s Apollo Go robotaxis stopped moving simultaneously. They did not pull over or park. They froze in place, …
Read on Medium →
11 April 2026
Your .docx Is Wasting 33% of Your AI Budget
The average knowledge worker produces documents in Microsoft Word, sends them as PDF, and occasionally pastes them into an AI tool for summarization or review. At each step, the file carries formattin…
Read on Medium →
9 April 2026
What Mythos Means: Elite Cyberattacks Just Got Much Cheaper
On April 7, 2026, Anthropic announced that it had built an AI model capable of finding and exploiting zero-day vulnerabilities (flaws unknown to the software’s own developers) in every major operating…
Read on Medium →
7 April 2026
The Code That Nobody Owns
When Anthropic accidentally shipped a 59.8 MB source map inside its Claude Code npm package on March 31, 512,000 lines of production TypeScript became public. Within hours, developer Sigrid Jin used O…
Read on Medium →
6 April 2026
Google Just Launched Gemma 4. Here’s What It Means for Your AI Budget.
A year ago, choosing an open-source AI model for production work meant accepting a meaningful gap in quality. You got privacy, you got control, and you got a model that was clearly worse than GPT-4 or…
Read on Medium →
5 April 2026
How to Install Your First Agent Operating System (and Why You Should)
Autonomous AI agents are not a future problem. The infrastructure layer for running them is being built right now, in public, by competing projects with very different ideas about security, architectu…
Read on Medium →
These are the 9 most recent. There are more on Medium.
View all articles on Medium →