Rev Lebaredian, NVIDIA’s Vice President of Omniverse and Simulation Technology, is leading the company’s ambitious push into AI-powered robotics and virtual simulation environments. Reporting directly to CEO Jensen Huang, Lebaredian’s mission is essentially to build “robot brains” that can learn and operate in the physical world.
Lebaredian’s career trajectory mirrors NVIDIA’s evolution from graphics technology to AI powerhouse. Starting in the movie industry where he helped bring characters like Mighty Joe Young and Stuart Little to life, he now applies similar principles to robotics simulation. For the past seven years, his focus has been on creating virtual environments where robots can safely learn and practice tasks before deployment in the real world.
At the heart of this effort is NVIDIA Omniverse, an artificial environment designed to be faithful to real-world physics and conditions. “The key here is that the simulation that we use has to match the real world as closely as possible so that what it learns inside that virtual world is transferable to the real world,” Lebaredian explained. This approach allows billions of repetitions and mistakes in a safe environment where no humans, robots, or property are harmed.
NVIDIA is following a proven playbook that worked with large language models (LLMs). The company built Megatron in 2021, one of the world’s largest models at the time, which Lebaredian says has “a direct line to GPT.” Now, NVIDIA is attempting to replicate that success with Project Groot, announced in March as a foundation model for robotics. Like their recent NVLM model released in September, these tools are licensed for research purposes only, not commercial use.
The strategy is deliberate: build the technology first, enable others to innovate on top of it, and wait for the industry to catch up. “We built all these computers. Nobody was asking for it, and everybody was skeptical,” Lebaredian said about their LLM work. The company is making similar long-term bets on robotics, investing heavily while waiting for the field’s “ChatGPT moment” to arrive. As Lebaredian notes, “AI is essentially bridging this computing to physical world divide, and that essentially is robotics.”
Key Quotes
AI is essentially bridging this computing to physical world divide, and that essentially is robotics.
Rev Lebaredian, NVIDIA’s VP of Omniverse and Simulation Technology, explains how artificial intelligence is enabling the convergence of digital computation and physical robotics, representing a fundamental shift in how machines interact with the real world.
The key here is that the simulation that we use has to match the real world as closely as possible so that what it learns inside that virtual world is transferable to the real world. If it learns inside a cartoon world with cartoon physics, it’s not going to behave properly when it gets into the real world.
Lebaredian emphasizes the critical importance of high-fidelity simulation in NVIDIA’s Omniverse platform, explaining why accurate virtual environments are essential for training robots that can safely and effectively operate in physical spaces.
There’s a direct line from Megatron, to GPT.
Lebaredian reveals NVIDIA’s foundational role in the generative AI revolution, connecting their 2021 Megatron model to the technology that eventually powered ChatGPT and the current AI boom, demonstrating the company’s pattern of early investment in transformative technologies.
I watched Jensen make these kinds of bets that are far-reaching, where there’s a lot of ambiguity as to when it’s going to happen or not, and just kind of navigate — believing on first principles that it’s going to get there.
Lebaredian describes CEO Jensen Huang’s strategic approach to long-term technology investments, explaining how NVIDIA positions itself ahead of major industry shifts by building infrastructure before market demand materializes, a strategy now being applied to robotics.
Our Take
NVIDIA’s robotics strategy reveals a sophisticated understanding of how technological revolutions unfold. By creating Project Groot and Omniverse as research tools rather than commercial products, NVIDIA is seeding an entire ecosystem—much like how their early GPU work enabled the deep learning revolution. The parallel to Megatron and GPT is particularly telling: NVIDIA doesn’t need to build the killer robot application; they just need to provide the foundational infrastructure that makes all robot applications possible.
What’s most striking is the simulation-first approach. While competitors race to deploy physical robots, NVIDIA is building the virtual training grounds where those robots will learn. This could prove decisive—whoever controls the best simulation environment may ultimately control the robotics industry. The stakes are enormous, and NVIDIA’s willingness to invest heavily while “not dying in the process” suggests they’re playing a long game that could reshape manufacturing, logistics, and countless other industries within the next decade.
Why This Matters
This story reveals NVIDIA’s strategic positioning at the intersection of AI and robotics, potentially the next major wave after generative AI. Lebaredian’s work on Omniverse and simulation technology represents a critical infrastructure layer for the emerging robotics industry, similar to how NVIDIA’s GPUs became essential for training large language models.
The implications are far-reaching: if NVIDIA successfully replicates its LLM playbook for robotics, it could dominate the physical AI market just as it dominates computational AI today. The company’s approach of building foundational tools and licensing them for research accelerates industry-wide innovation while establishing NVIDIA as the de facto standard.
For businesses and society, this signals that practical, AI-powered robots may be closer than expected. The simulation-first approach solves one of robotics’ biggest challenges—safe, scalable training—potentially accelerating deployment across manufacturing, logistics, healthcare, and service industries. The fact that NVIDIA is making massive investments “without dying in the process” demonstrates confidence that robotics will follow generative AI’s explosive growth trajectory, fundamentally transforming how physical work is performed.
Recommended Reading
For those interested in learning more about artificial intelligence, machine learning, and effective AI communication, here are some excellent resources:
Recommended Reading
Related Stories
- Jensen Huang: TSMC Helped Fix Design Flaw with Nvidia’s Blackwell AI Chip
- Tech Workers Are the Real Winners in the AI Talent War, With Pay Set to Soar by 2024
- Pitch Deck: TensorWave raises $10M to build safer AI compute chips for Nvidia and AMD
- EnCharge AI Secures $100M Series B to Revolutionize Energy-Efficient AI Chips