Photo credit: venturebeat.com
Rescale, a digital engineering platform specializing in complex simulations and calculations in the cloud, has announced it secured $115 million in Series D funding. This financial boost will enhance the development of AI-driven engineering tools designed to significantly accelerate product design and testing processes.
This funding round elevates Rescale’s total capital raised to over $260 million and includes investment from notable entities such as Applied Ventures, Atika Capital, Foxconn, Hanwha Asset Management Deeptech Venture Fund, Hitachi Ventures, NEC Orchestrating Future Fund, Nvidia, Prosperity7, SineWave Ventures, TransLink Capital, the University of Michigan, and Y Combinator.
Founded in San Francisco, Rescale has garnered support from high-profile early backers, including Sam Altman, Jeff Bezos, Paul Graham, and Peter Thiel. This latest funding round aims to advance Rescale’s mission to revolutionize product design across various sectors by leveraging high-performance computing and intelligent data management as well as what the company terms “AI physics.”
“Rescale’s founding mission centers on empowering engineers and scientists to hasten innovation through efficient computations and simulations,” explained Joris Poort, founder and CEO of Rescale, in a conversation with VentureBeat. “This focus remains at the heart of our efforts.”
From Boeing’s Carbon Fiber Challenge to a $260 Million Startup
The genesis of Rescale reflects Poort’s prior work on the Boeing 787 Dreamliner over two decades ago, where he and co-founder Adam McKenzie faced the challenge of designing the aircraft’s wing using complex physics-based simulations.
“Adam and I were at Boeing, running substantial physics simulations for the 787,” Poort recounted. “As the first fully carbon fiber commercial airplane, it presented numerous engineering challenges. Unlike aluminum, carbon fiber comprises various layers and factors that required optimization.”
During that time, they struggled with inadequate computing power, which made it impossible to execute the numerous calculations necessary for the cutting-edge carbon fiber designs. “We faced a scarcity of compute resources back then, long before cloud computing emerged,” he recalled. “We had to piece together capabilities from different organizations just to conduct those extensive simulations over a weekend.”
This predicament inspired the founding mission of Rescale: to create the platform that Poort and McKenzie wished they had during their Boeing tenure.
“We created Rescale to fulfill the need for the tools we lacked, as it took years to develop the required capabilities,” he noted. “We were engineers striving to design the optimal aircraft, yet we transitioned into applied mathematicians and computer scientists, focusing on the infrastructure challenges instead of pure engineering.”
How AI Models are Transforming Days of Calculations into Seconds
A pivotal aspect of Rescale’s strategy is the concept of “AI physics.” This approach employs AI models trained on simulation data to significantly reduce computation times in engineering. Traditional physics simulations can span several days, yet AI models trained on similar datasets can yield approximate results in seconds.
“With the introduction of AI physics, we train models on extensive simulation data sets to accelerate computations by over 1,000 times,” Poort stated. “While AI offers probabilistic solutions—essentially estimates—traditional calculations provide deterministically precise outcomes.”
For example, Rescale has collaborated with customers like General Motors, who utilize AI to optimize the aerodynamics of a Formula One car. Previously, the requisite aerodynamic calculations might take three days even on a thousand compute cores, but with AI, they can achieve this in under a second.
This remarkable acceleration equips engineers to explore a broader array of design iterations rapidly, enhancing design flexibility and innovation.
“The distinct advantage of AI physics is verifiability. It’s fundamentally mathematics,” Poort asserted. “Unlike large language models, which sometimes generate unverifiable outputs, physics offers concrete solutions that can be validated.”
The recent funding aligns with a broader trend of increased investment in technologies that enhance product development, as the high-performance computing sector has ballooned to an estimated $50 billion, with simulation software contributing about $20 billion and product lifecycle management around $30 billion, as highlighted by Rescale.
Rescale’s uniqueness stems from its innovative “compute recommendation engine,” designed to optimize workloads dynamically across various cloud platforms.
“Our compute recommendation engine allows real-time optimization of workloads across diverse public cloud architectures,” Poort elaborated. “We support over 1,150 applications across numerous versions, operating systems, and hardware architectures, resulting in a staggering 50 million unique configurations.”
Enterprise clients, including Arm, General Motors, Samsung, SLB (formerly Schlumberger), and the U.S. Department of Defense, collectively invest over $1 billion yearly in high-performance computing infrastructure.
Beyond Simulation: Data Management and AI Integration for Modern Engineering
Rescale is actively evolving its roadmap across three primary areas: expanding its repository of more than 1,250 applications and network of over 500 cloud data centers, enhancing unified data management and digital thread capabilities, and driving engineering efficiencies through AI.
“We’ve also developed a product named Rescale Data, focusing on establishing an intelligent data layer,” Poort shared. “This encompasses what’s often referred to as the digital thread. Whether in aerospace, automotive, life sciences, or pharmaceuticals, it is crucial to track data throughout the product lifecycle. Should an issue arise, it allows for reviewing data creation timelines, input files, and connected information.”
One of the investors in this funding round, Applied Materials, has been collaborating with Rescale to refine simulation capabilities. The partnership indicates a deeper shift in the methodologies for capturing and applying engineering knowledge, rather than merely expediting existing processes.
Significantly, Rescale’s methodology integrates traditional physics simulations and AI approximations, establishing a more secure pathway for AI’s incorporation into precision-critical fields. The clarity of mathematical verification in engineering offers a contrast to the subjective validation challenges typical in language models.
Though Poort acknowledges that the vision of a “foundational physics model” — a sophisticated AI trained on extensive physics data capable of discovering new principles — is still aspirational, Rescale is dedicated to delivering practical benefits today through targeted AI applications.
“When it comes to quantum technology, it remains in an early commercialization phase,” Poort noted, differentiating from his focus. “From Rescale’s perspective of AI physics, our emphasis is on concrete problems that AI advancements can resolve for immediate ROI.”
What Silicon Valley Luminaries See in Engineering Simulation
Rescale has drawn insights from prominent early investors to guide its strategic direction. Paul Graham, who was the first investor in the firm, continues to impart wisdom regarding company culture. Sam Altman provides valuable insights into AI-related infrastructure, while Jeff Bezos offers perspectives from his work with Blue Origin. Peter Thiel shares advice on scaling ventures and establishing government collaborations.
“Paul Graham’s support has been instrumental—it was he who first recognized our potential. His essays are a wellspring of wisdom, and I still consult with him regularly,” Poort shared with VentureBeat. “Sam Altman has been an advocate since our inception; his expertise in AI and technology is invaluable.”
On Bezos, Poort added, “His insights from Blue Origin provide a unique perspective on how aerospace firms can utilize cloud technology, an area we initially focused on.”
“Having long-term visionaries among our partners is crucial,” Poort articulated. “Jensen’s profound understanding of technical elements and future technology trends is another key asset.”
Engineering’s Cloud Transition Offers New Possibilities
As geopolitical factors increasingly shape sectors like semiconductors and defense, Rescale is strategically positioned to assist clients in managing regulatory challenges and shifting supply chains. The company has honed its capabilities to integrate with sovereign clouds—region-specific cloud infrastructures that retain data within national boundaries.
“We are witnessing the rise of sovereign clouds,” Poort explained. “As various nations create tailored cloud infrastructures for specific applications, our strategy involves collaborating with these providers to align services with customer needs. For instance, we can support a Japanese client operating exclusively on a domestic cloud.”
With fewer than 20% of high-performance computing resources currently transitioned to the cloud, Rescale sees enormous potential for growth as more engineering tasks move to cloud environments. Their innovative AI physics methodology could change the landscape of product design across sectors, leading to reduced development timelines and enhanced performance.
“A critical insight is that with adequate computing resources, we can achieve markedly improved designs,” Poort reflected on his Boeing experiences. The $115 million investment signifies a growing confidence that the divide between imaginative design and technical feasibility is narrowing—not through revolutionary physics advancements, but by smarter utilization of existing data and simulation techniques.
As for the Boeing 787 wing that inspired all of this? “If you’ve had the chance to fly on a 787 Dreamliner,” shared Poort, “you’ll recognize its distinctively curved wing design, a testament to our work that has led to a 20% increase in fuel efficiency for the aircraft.”
Source
venturebeat.com