Agents in Practice
Putting Claude Agents to Work: Automating the Mundane
The highest-ROI use of AI agents isn't replacing your best people—it's deleting the repetitive, low-judgment work that quietly drains your team every week.
Read article →
From the blog
Field notes on autonomous agents, frontier AI, and the discipline of running it in production.
Frontier AI
World Models: The Next Frontier in Video Generation
Video models are quietly becoming world models—learning physics, permanence, and cause-and-effect. Here's what that shift means for the businesses building on top of them.
Read article →
Governance
Treating AI Agents as Production Software
Anyone can add a bot to Slack. Making it safe, reliable, and measurably useful is an engineering discipline—not a prompt. Here's the operating model that holds up.
Read article →
Architecture
MCP and the Tool-Use Era: Giving Agents Safe Access
An agent is only as useful as the systems it can reach. The Model Context Protocol is quietly becoming the standard way to connect agents to your tools—safely.
Read article →