The price of using machine learning intelligence is decreasing by several orders of magnitude a year and shows no signs of slowing. We are likely to see LLMs integrated into every part of computing as infrastructure, not just as a product, in the next year or two. We are likely to see LLMs running locally at current frontier-quality on commodity hardware in the next 3-6 years. Starting very soon, we are likely to see quality and access become the limiting factor to AI use, not sheer number of tokens.

Last week the frontend and mobile dev community sat up when Shopify announced it was moving its mobile apps off React Native and back to Swift and Kotlin. Shopify went all-in on React Native in 2020, and as recently as January 2025 Shopify’s head of mobile Mustafa Ali wrote that its future looked bright and that Shopify would keep investing in it. Not quite two years later, the company has reversed course.

I ran five different coding agents at the same local model, on the same task, with the same frozen test suite — and then I counted why they failed. The answer wasn’t subtle. About 90% of the failures were harness problems. Only about 10% were the model.

When writing object oriented code, a pretty useful technique is delegation–letting one object figure things out by sending messages to another object.