When AI Makes Billions or Fails: 5 Projects from Recent Years and a Football Bonus

At Techmates we live for code, architecture and innovation. We do not just follow mainstream headlines, but we love to look under the hood of projects that showcase uncompromising engineering craftsmanship, brutal hardware optimization or useful systems integration.

We selected five technological achievements from recent years that genuinely caught our attention, and on top of that we are adding one special and highly current sports bonus.
‍

Halter or "Cowgorithm" in Real Practice

The idea of connecting the biology of a living creature with artificial intelligence and the cloud sounds like science fiction. However, the New Zealand startup Halter turned this into a two billion dollar business. They developed solar powered collars for cows that use advanced algorithms to remotely manage the entire herd. The system works on the basis of gentle vibrations and audio signals, allowing the farmer to move animals remotely via an app and create virtual fences without the need to build physical structures. For us at Techmates, this is a perfect example of how massive data collection from IoT devices meets smart analytics in the cloud.

Lessons from Ikea: Do You Want to Save or Make Money with AI?

While half of all companies see AI only as a tool for laying off people and mindlessly cutting costs, the Swedish giant IKEA put on a masterclass in business. They deployed a chatbot that now resolves 57% of customer inquiries, freeing up the capacity of 8500 call center operators. Instead of firing them, the company retrained them into interior design consultants who started creating 3D renderings and floor plans via video. The result? They saved 13 million euros on the chatbot, but the new design channel generated an additional annual revenue of 1.3 billion euros. This is exactly how we think about automation at Techmates. AI should not replace people, it should take away their boring and repetitive monkey work so they can focus on creative tasks with real value.

‍

Coca Cola and Shortening Decision Making from 14 Days to 1 Hour

While the creative attempts of Coca Cola with AI generated ads faced a lot of criticism, the beverage giant pointed in exactly the right direction behind the scenes with hard data. The current retail market is extremely fragmented. Half of the people are cutting back severely, while the other half demands premium products. Coca Cola therefore let AI instantly analyze customer behavior right at the store shelves. Within hours, the system determines where to send cheaper packaging and where to support margins with a premium portfolio. They shortened the decision cycle from two weeks to just a single hour. They do not command algorithms to have human emotion, they demand brutal speed in analytics. This confirms our long term experience: The best AI in a company is often invisible to the customer, but it allows you to react to market changes first.

‍

Rankacy from Ostrava: From Counter Strike 2 to Real Business

This local project from Ostrava absolutely thrilled us at the ML Prague conference. The startup Rankacy started as an analytical tool for professional players of Counter Strike 2. It processed data from millions of players and created its own predictive models that analyze behavior and make lightning fast decisions based on in game mistakes. The Rankacy team managed a great feat, they took this mathematical foundation and started bending it for the real world. They demonstrate that if AI can learn to analyze complex behavior in a tactical video game, it can apply the same principles to industry, logistics or defense. For us, this is a brilliant example of pure data engineering that built a global technological asset from scratch.

‍

When You Overdo It: Voice Drive Thru by McDonalds and IBM

Not every project ends in success, and even giants sometimes hit reality hard. McDonalds spent a long time testing an automated voice cashier powered by IBM technology. The goal was to completely robotize drive thru orders. However, the experiment at a hundred locations recently came to an end after several years of testing. The system failed to reliably filter out engine noise, kids screaming in the back seats and various human accents. The results were bizarre orders like bacon ice cream or nine iced teas instead of one, which flooded social media. For us, this is a clear message. Voice AI has a massive future, but without perfect data cleaning and mastering the real world context, it will bring more harm than benefit.

Current Bonus: Trionda or AI on the Football Pitch

The biggest technological bombshell outside the pure development world for us is the official ball for this year's 2026 Football World Cup called Adidas Trionda. Forget an ordinary piece of leather, this is currently the most modern piece of hardware connected to cloud systems.

Inside the ball, opposite the valve, is a special sensor (IMU) that tracks position and movement in 3D space 500 times per second. The sensor runs on an integrated battery and sends data in real time directly to the video assistant referee (VAR) system.

The artificial intelligence at the stadium then instantly combines this data with camera footage tracking 29 points on the body of each player. The result is a semi automated system for offside tracking and immediate detection of handball offenses or ball trajectories. Furthermore, the data allows TV networks to instantly generate 3D animations for viewers.

Techmates: Your Data Under Your Roof

All these selected projects clearly show that true business success or hitting a hard wall of reality depends on mastering technical details one hundred percent. As engineers and developers at Techmates, we help you bring similarly powerful innovations directly into your practice by taking advanced open source models and integrating them into your existing ERP and CRM systems via stable API interfaces. We will help you correctly set up the entire data architecture so that you minimize risks and keep the whole ecosystem fully under control in your own private environment.

The examples of Ikea or Coca Cola clearly demonstrate that smartly deployed AI in the background can generate billions and save precious time. At the same time, we know that without precise engineering, it ends up as an expensive misstep. At Techmates, we build solutions that make business sense and actually work. If you want to see our results, check out our projects, or launch a new project straight away with us.

‍