- Waymo is the biggest agentic product in the world, and it shows what depth actually costs. Replacing one human, the driver, took tens of billions of dollars of edge-case training and data that exists nowhere on the internet. You cannot drop the next Anthropic model into your Mercedes and tell it to drive you home. Every enterprise agent that truly replaces a person needs that same depth, which is why most agentic enterprise products are not ready x.com/gokulr/status/2071692278582890889
- Good engineering has traditionally meant shrinking the space of possible outcomes until the same action reliably produces the same result. AI complicates that logic: much of its value comes from finding paths we didn’t explicitly define. So instead of prescribing exactly what it should do, we define the boundaries within which it can find its own way linear.app/now/how-we-built-linear-agent
- With sufficient AI progress, the interest on your tech debt is sub-zero. How should you behave if you believe that to be true? You should probably spend less time worrying about tech debt, and spend more time shipping new features. Many companies are doing just that. But it’s a gamble. We don’t know how much better AI will get, nor whether it can improve fast enough to undo the mess that was made in anticipation of its improvement jesseduffield.com/Negative-interest-tech-debt
- Stripe acquired OpenRouter to aggregate inference demand and give developers an easier way to manage the intelligence pipeline, just like how they made it easier for developers to manage money stratechery.com/2026/stripe-acquiring-openrouter-aggregating-ai-flipping-the-business-model and x.com/ericnewcomer/status/2090133049291788434
- The more people you consult for advice, the more likely you’ll make a bad or average decision. Because people on average are risk averse, may hate opportunities they lack, and tend to mix cynicism and criticism x.com/scottbelsky/status/2087283610952855777
What I’ve been reading (W34, 2026)
What Waymo says about agent depth, defining boundaries instead of steps, the tech debt gamble, Stripe buying OpenRouter, and why crowdsourced advice averages down