Discover how Nvidia’s edge AI technology powers ADAM, a robotic bartender serving fans at Las Vegas’s T-Mobile Arena, transforming hospitality automation.Robotics is rapidly reshaping the hospitality industry as businesses look for innovative ways to address ongoing labour shortages. What began with back-of-house automation is now moving directly into guest-facing experiences—and a Las Vegas sports arena is leading the way.
At T-Mobile Arena, home of the NHL’s Vegas Golden Knights, fans are being served drinks by ADAM, a dual-arm robotic bartender powered by Nvidia’s edge AI technology. The deployment highlights how intelligent automation can deliver both operational efficiency and memorable customer interactions.
Meet ADAM: The Automated Dual-Arm Mixologist
Short for Automated Dual Arm Mixologist, ADAM is a real-world example of edge AI computing in action. Developed by Las Vegas-based Richtech Robotics, the robot was designed to support hospitality teams while enhancing the fan experience during live sporting events.
Built using Nvidia’s Isaac robotics libraries, ADAM combines precision, adaptability, and consistency—qualities that are increasingly critical in high-volume service environments.
“The hospitality industry continues to face serious staffing challenges, and ADAM was created to meet those needs while elevating the guest experience,” says Matt Casella, former President of Richtech Robotics. “Powered by Nvidia’s Isaac platform, the system is scalable, reliable, and creates moments people remember.”
Training in a Virtual Bar Before Serving the Crowd
Before pouring its first drink, ADAM underwent extensive virtual training. Using Nvidia Isaac Sim, a robotics simulation framework built on Nvidia Omniverse, Richtech Robotics created a digital twin of the robot’s bar environment.
This simulated setup replicated real-world variables such as lighting changes, reflective surfaces, cups, and bar tools. By training in this virtual environment, ADAM learned to handle visual challenges it would later encounter during live service.
The approach generated synthetic data—artificially created training inputs that help AI systems recognise objects under diverse conditions. This proved essential for teaching ADAM to accurately identify items despite glare, reflections, or changing surroundings.
How Edge AI Powers ADAM’s Precision
ADAM’s operational intelligence was refined using Isaac Lab, Nvidia’s open-source robot learning framework. This enabled the robot to master complex actions such as pouring, shaking, and adjusting grip pressure—allowing it to respond dynamically rather than follow rigid programming.
At the core of the system is Nvidia Jetson AGX Orin, an edge AI computing platform capable of delivering up to 275 trillion operations per second. By processing data locally instead of relying on the cloud, ADAM achieves ultra-low latency and real-time responsiveness—essential for fast-paced service environments.
Real-Time Perception and Control
Using Isaac ROS 2 libraries, ADAM continuously processes camera inputs to detect objects and recalibrate its workspace in real time. Its perception system—developed with Nvidia’s TAO Toolkit and optimised using TensorRT—can recognise cups, monitor liquid levels, and adjust movements with latency below 40 milliseconds.
This allows ADAM to detect misplaced cups, stop pours when foam reaches the rim, and make instant corrections without disrupting service.
Beyond the Bar: Robotics for Industry
Richtech Robotics is extending this technology beyond hospitality. At Nvidia’s GTC DC conference, the company unveiled Dex, a mobile humanoid robot designed for factories and warehouses.
Dex combines a wheeled autonomous base with dual-arm manipulation to perform tasks such as machine operation, parts sorting, and material handling. Powered by Nvidia Jetson Thor, the system was trained using a combination of real-world data and synthetic simulations generated through Isaac Sim.
“The response at T-Mobile Arena has been phenomenal,” Casella adds. “People genuinely enjoy interacting with ADAM—and that’s where robotics is headed: practical, intelligent, and engaging.”
Read more: How GPT-5.2 and FrontierScience Are Pushing AI Toward Real Scientific Breakthroughs