In Las Vegas, fans at the T-Mobile Arena are experiencing more than just hockey. They’re getting a glimpse into the future of hospitality with ADAM, a robot bartender developed with NVIDIA Isaac libraries. ADAM is serving drinks and turning heads, bringing cutting-edge AI to one of the NHL’s most exciting venues.
ADAM, which stands for Automated Dual Arm Mixologist, was created by Richtech Robotics, a Las Vegas-based company. It’s more than just a novelty; it’s a solution to pressing challenges in the hospitality industry, including labor shortages and growing demand for unique customer experiences.
The Future of Hospitality with ADAM
“The hospitality industry faces significant labor challenges, and ADAM is our answer to meeting those needs while elevating the customer experience,” said Matt Casella, president of Richtech Robotics. He adds, “With NVIDIA’s Isaac platform, we’ve developed a solution that’s scalable, consistent, and creates memorable moments for fans. The response at T-Mobile Arena has been phenomenal—people love interacting with ADAM.”
Read Also
First Game Made Entirely with Generative AI Now Has a Demo
Rivian Builds In-House AI Assistant Ahead of December Reveal
Trump’s Executive Order on AI Regulation: Why It Hits California the Hardest
Training ADAM to Serve Drinks in Simulation
Before ADAM served its first drink, it trained in a virtual environment. Richtech used NVIDIA Isaac Sim, an open-source robotic simulation framework built on NVIDIA Omniverse, to create a high-fidelity simulation of ADAM’s workstation. The simulated bar included cups, utensils, and varying lighting conditions to ensure ADAM could recognize objects even under challenging circumstances like glare or reflections.
The team also used Isaac Lab, NVIDIA’s open-source robot learning framework, to fine-tune ADAM’s skills. The result is a robot that adapts to its environment with precision, pouring and shaking drinks to perfection.
ADAM Runs on NVIDIA Jetson for Real-Time AI
ADAM operates on NVIDIA Jetson AGX Orin, a powerful edge AI platform capable of 275 TOPS of compute. With Isaac ROS 2 libraries, ADAM captures real-time camera feeds, detects objects, and calibrates its workspace with less than 40 milliseconds of latency. This enables ADAM to adjust its movements, correct missteps like misplaced cups, and ensure a perfect pour every time.
ADAM’s AI-powered perception system, optimized with TensorRT and the TAO Toolkit, allows it to detect when foam reaches the rim of a cup or when a pour is off balance—adjusting quickly without missing a beat.
Dex: AI Meets Industrial Automation
While ADAM is making waves in the hospitality sector, Richtech Robotics is also advancing industrial automation with Dex, a new mobile humanoid robot designed for factory and warehouse settings. Unveiled at GTC DC, Dex combines the mobility of an autonomous wheeled platform with the precision of dual-arm dexterity. It’s built to handle tasks like machine operation, parts sorting, and packaging, making it adaptable to different industrial workflows.
Dex also runs on NVIDIA Jetson Thor, a next-gen robotics processor that provides real-time sensor processing and AI reasoning in dynamic environments. Trained with a blend of real-world and synthetic data from Isaac Sim, Dex’s model is generalized to perform across various industrial tasks.
