Big week in AI! We're seeing a full-on race to build the best coding agents, with Anthropic and OpenAI both dropping major upgrades to Claude Code and Codex. We'll break down what "Coordinator Mode" means and why everyone's suddenly obsessed with parallel sub-agents.

Plus, we're diving into the wildest story of the week - Anthropic allegedly sitting on their most powerful model because it's too good at hacking. We'll also talk about why AI companies might be pricing themselves out of the market (spoiler: nobody wants $7 Doritos), and what the future of software looks like when it's built for agents, not humans.

Grab your coffee, this one's packed with insights you need to know.