Eat Smart: A smart fridge instead of a company canteen
Chef Filip Sajler and HOPI Holding wanted to bring restaurant-quality food into companies without building canteens. We built the entire technology from scratch - RFID recognition, mobile app, cloud. Today the network runs on 80 locations and sold 312,000 portions in 2025.
EAT SMART answers a question almost every company asks: how to feed employees quality food without building and running an expensive canteen. Chef Filip Sajler and HOPI Holding came up with a self-service smart fridge stocked with fresh chef-made meals. We supplied the technology behind the idea - a complete ecosystem from hardware through mobile app to cloud that turns the whole grab-and-go moment into a 15-second purchase.
“If it weren't for Cookielab, we might have scrapped the project long ago. But Radek, Kolda and their team have partnered up with us to push the product forward. They don't shy away from challenges and they understand that cooperation isn't only about development, but also about presenting to investors.”
The story
Filip Sajler and HOPI Holding are no newcomers to corporate dining - they had already built a canteen network and sold it to the American group Aramark. With EAT SMART they set a more ambitious goal: workplace dining that scales without chefs and staff on site. Instead of tens of millions of crowns for building a canteen and millions more each month to run one, a company gets a smart fridge for tens of thousands of crowns a month - up to 90% lower investment and 75% lower operating costs.
We started with a pilot at the HOPI campus at the end of 2023; the first live launch came in January 2024. Since then we have kept expanding the product: Apple Pay and Google Pay, a credit and discount system for marketing campaigns, dynamic in-app content, remote tablet management, and tools for operations and production. The product is also getting smarter inside - an AI pipeline that enhances meal photos is already deployed, an AI agent for checking customer claims runs as a functional prototype, and demand prediction is in development.
The growth speaks plainly: around 30 locations at the end of 2024, 80 by the end of 2025, and 312,000 portions sold that year - three times the year before. Customers include WITTE Nejdek, Fehrer Bohemia, T-Mobile, and Moneta Money Bank.
The challenges
This was never a pure software project. It combines physical hardware, fresh food with a short shelf life, and software that has to run reliably 24/7. The hardest fight was RFID itself: the chips initially failed because metal parts of the fridge and the conductive black paint interfered with the signal. For a while it looked like physics might win. Fixing it meant completely re-engineering the reading array around specialized LOCFIELD UHF antennas.
What's next
EAT SMART is heading beyond the Czech Republic and plans to roughly double the network to 160 fridges in 2026. On our side, we are working on AI demand prediction that should cut write-offs of uncollected meals, and on further simplifying the day-to-day operations of the network.
From an engineering perspective, EAT SMART is a distributed IoT system where edge computing, RFID hardware, mobile payments, and cloud meet. Every fridge is effectively a standalone point of sale that has to keep selling through connectivity outages and know precisely what a customer took. That means solving problems a typical app never hits: radio interference from metal, syncing edge transactions to the cloud, and managing a fleet of devices in the field.
What we delivered
“EAT SMART is a project where software meets the real world - metal, cold, radio waves, and hungry people. Exactly the kind of challenge we enjoy.”
Technologies
The story
The architecture stands on AWS IoT. Each fridge runs as an edge node on AWS Greengrass with application logic in Node.js and TypeScript, so detecting a pickup and opening a transaction works locally even without a stable connection, and data syncs to the cloud afterwards. Above it, AWS IoT Core and Lambda process transactions, stock levels, and telemetry such as temperature and door state.
RFID is the heart of the experience. Every box carries a UHF chip; when the door closes, the fridge reads what is left and derives what was taken from the difference. Reliable reading cost us a complete rework of the antenna array - metal reflected and attenuated the signal, graphite in the black paint turned out to be electrically conductive, and adhesives caused interference. LOCFIELD UHF antennas with a FEIG controller finally solved it. The pilot added one more lesson: the people who cook and pack the meals are part of the system, because a tag placed wrong is a tag the antenna cannot read.
The mobile app is native iOS and Android; the on-fridge terminal is a React webview in kiosk mode, managed remotely via MDM. The operations team runs on Retool tools for restocking, monitoring, and fixing transactions without a developer.
AI and automation enter the project in two ways: in the product and in how we build it. Deployed today: the automated image-enhancement pipeline for meal photos and ChefTech's real-time error detection on the production line. A functional prototype: an AI agent for checking customer claims. In development: ML demand prediction to minimize meal write-offs. In our own workflow, automated code reviews with CodeRabbit and Claude Code keep delivery fast on a long-running codebase.
What's next
We are working on AI demand prediction to cut write-offs and continuing work on the claims-checking prototype. In parallel, we are optimizing fridge hardware costs during the move to a new supplier and preparing the platform for a network of around 160 fridges in 2026.
