I have a ChatGPT thread where I send screenshots of grocery receipts and have AI compare prices across stores and delivery services. Which store is cheaper? How much extra am I paying for delivery? Is going in person actually worth it? I find this stuff interesting but the differences are not enough to reorganize my life around. Except one. Yogurt. My kids love yogurt and the ones they eat are 40% cheaper at Costco. The yogurt-nomics alone are enough to justify shopping mostly at Costco even with delivery.

Evals have been coming up more and more in my conversations with podcast guests and PM friends. Nearly half of the 25 awesome PM job openings I shared last week ask for experience writing evals. And leading companies keep sharing what investing in evals bought them: — @tryramp took its automatic receipt collection from 35% to 83% accuracy. — @Shopify shipped an AI workflow builder that's 2.2x faster and 68% cheaper than the frontier-model setup it replaced. — @harvey__ai rebuilt its AI contract reviewer, nearly doubling its internal quality score. — @cursor_ai tuned its Auto Balance routing, with much higher user satisfaction at 41% lower cost. So I asked the 🐐s of evals, @HamelHusain and @sh_reya, to write an advanced sequel to their very popular "Building eval systems that improve your AI product." Drawing on their work with 50+ AI companies, they share the key step most teams skip, what you should (and shouldn't) automate, and a free plugin that lets a coding agent do most of the heavy lifting. Read it here:

Introducing Gemini 4 Argon – our new frontier model. It’s built for complex workflows across coding, enterprise knowledge work, and cybersecurity defense – rolling out today to a set of trusted testers through our Fairwind Program.

GPT-5.6 Sol, Luna and Terra are listed as "Unlimited*" usage on the Pro plan page. It looks like OAI are giving unlimited (*with fair usage) use of those models to balance out this change in pricing. 5.6 Sol and Luna X-High are two of my fave models so this could be great!