ZingZee

😎

Monday morning in a business that runs on its own system looks like this. Who is checking in today? Three guests, three welcome messages sent in one tap. Pay the cleaners for the week? The rota is already there, four payments go out, the receipt comes back. Is the pool chlorine low? The order is placed, the van is on the map, the stock bar refills when it arrives. How is the month looking? Every number is live. ZingZee builds platforms like this for hospitality, retail, logistics and professional services companies, and connects them to what those companies already use: SAP, Salesforce, WhatsApp, Stripe, JCC, Xero, Booking.com and the rest. Every project runs through the same five phases, assess, design, build, test, run, so the scope, the cost and the hand-over are clear before the work starts. Based in Limassol, engineers across Europe. zingzee.com

The price of intelligence fell again this week, and the ceiling went up. Anthropic shipped Claude Fable 5.1 on Monday. Same price as Fable 5, about 25 percent cheaper to run in practice, up to 45 percent on agent workloads, because cached reads dropped to 25 cents per million tokens. OpenAI's GPT-6 Astra started rolling out yesterday. Strong enough at finding software vulnerabilities that OpenAI is gating its full version to a vetted security program first. Z.ai put out an open 1M-context model at 15 cents per million tokens in. Google is running Gemini 3.8 Flash at half price until the end of the year. Eighteen months ago the question was whether a normal business could afford to run this. This week the question is which of your workflows you are still paying people to do by hand. We are a software development company. We build the systems those workflows live in, and since this summer we have an AI division that wires the models in, properly, with your data staying yours. If you want to know which three things in your business would pay for themselves first, message us. zingzee.com #SoftwareDevelopment #AI #BusinessDevelopment

The price of intelligence fell again this week. OpenAI cut GPT-5.6 API prices 20%. Anthropic cancelled the Sonnet 5 price rise. A 27B open model from Alibaba now runs offline on a single workstation. Alibaba raised 10.2 billion dollars in Hong Kong to build its own stack. Eighteen months ago we were arguing about whether this stuff was affordable to run inside a normal business. 🤷🏻