Cheers to AI: ADAM Robot Bartender Makes Drinks at Vegas Golden Knights Game

Source: NVIDIA AI Blog

ADAM at T‑Mobile Arena

In Las Vegas’s T‑Mobile Arena, fans of the Golden Knights are getting more than just hockey — they’re getting a taste of the future. ADAM, a robot developed with NVIDIA Isaac libraries, is pouring drinks and turning heads in one of the NHL’s most exciting venues.

ADAM, short for Automated Dual Arm Mixologist, was developed by Las‑Vegas‑based Richtech Robotics. It’s not just a novelty — it’s a solution to real‑world challenges in hospitality: labor shortages and demands for unique customer experiences.

“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. “With NVIDIA’s Isaac platform, we’ve developed a solution that’s scalable, consistent, and frankly, creates memorable moments for fans. The response at T‑Mobile Arena has been phenomenal—people love interacting with ADAM.”

Video: Watch the robot in action

Learning to Serve Drinks in Simulation

Before ADAM ever poured a drink, it trained in a virtual bar. Richtech used NVIDIA Isaac Sim, an open‑source, reference robotic simulation framework built on NVIDIA Omniverse, to build a high‑fidelity and physically accurate simulation of ADAM’s workstation, complete with cups, utensils and lighting variations. The team generated synthetic data to teach ADAM how to recognize objects even in tricky conditions like glare or reflection.

ADAM’s skills such as pouring and shaking were refined in simulation using Isaac Lab, NVIDIA’s open‑source robot learning framework. The result: a robot that doesn’t just follow instructions — it adapts to its environment with precision.

Running Real‑Time AI at the Edge With Jetson

ADAM runs on NVIDIA Jetson AGX Orin, the powerful edge AI platform capable of 275 TOPS of compute. Using Isaac ROS 2 libraries, ADAM captures camera feeds, detects objects and calibrates the workspace in real time. ADAM’s perception stack — built with the TAO Toolkit and optimized with TensorRT — enables it to identify cups, measure liquid levels and adjust movements with less than 40 ms of latency.

That means ADAM can spot a misplaced cup, detect when foam reaches the rim and correct a pour — all without missing a beat.

Creating Industrial Dexterity With NVIDIA Thor

While ADAM is busy serving drinks at Golden Knights games, Richtech Robotics is also making major strides in industrial automation with Dex, a new mobile humanoid robot built for factory and warehouse environments.

Recently unveiled at GTC DC, Dex combines the mobility of an autonomous wheeled platform with the precision of dual‑arm dexterity. It’s designed to handle light‑to‑medium industrial tasks such as machine operation, parts sorting, material handling and packaging — all with the flexibility to take on different tools and workflows.

Dex runs on NVIDIA Jetson Thor, a next‑generation robotics processor that delivers real‑time sensor processing and AI reasoning in dynamic industrial settings. Dex was trained from a blend of real‑world and synthetic data generated from Isaac Sim, allowing its model to generalize across a multitude of scenarios.

Learn more about Jetson Thor.