Legal name: Fainite AG
Fainite Deep Dive: A Zurich Physics-AI Simulation Startup Whose Caltech, ETH Zurich, Cambridge and Google Team Uses Physics-Aware Deep Learning to Cut Simulation from Tens of Hours to Seconds, With Only Venture Kick Grants and No Priced Round
A Zurich physics-AI simulation startup founded in 2025 by a Caltech, ETH Zurich, Cambridge and Google team; its Physics AI Engine uses physics-aware deep learning to cut simulation runs from tens of hours to seconds and ships an agentic Copilot for conversational setup, yet holds only Venture Kick grants of roughly CHF 190k and has disclosed no priced round or revenue.
Executive Summary
Fainite AG is a Swiss physical AI startup founded in 2025 in Zurich, building an engineering AI platform that turns simulation into inference. The founding team comes from Caltech, ETH Zurich, Cambridge and Google: Alex Donzelli as CEO with a Caltech master's in applied mechanics and prior roles at Saipem, Optotune and UBS, Professor Burigede Liu as chief scientist from Cambridge with earlier posts at Caltech and ETH, Matthias Bonvin as machine learning lead from ETH, Jan Pfeifer as machine learning scientist after twelve years at Google as a core contributor to TensorFlow Lattice, Ranking and Decision Forests, Roger Keene in commercial strategy after serving as Dassault SIMULIA worldwide vice president of sales and operations, and Professor Dennis Kochmann as technical advisor from ETH.
On funding the only disclosed capital is 150,000 Swiss francs from Venture Kick in 2026, about 170,000 euros, with no priced round, valuation or revenue disclosed. That places the company between technical validation and first industrial applications rather than at scale.
The product is a Physics AI Engine positioned to accelerate, simplify and automate simulation workflows, combining physics-aware deep learning models trained on customer CAE data, an AI agent guiding complex setup with expert recommendations, and a reusable results library. The company contrasts conventional workflows at 48 hours or more against 10 seconds with Fainite, and lists three customer pains: computation above ten hours, workflows spread across more than ten tools and simulation knowledge trapped in thousands of files.
On market framing the company cites 17 million engineers worldwide on its careers page and 9 million hardware engineers affected by simulation bottlenecks in the Venture Kick release, a 27 billion dollar simulation software market doubling within five years at 10 to 13 percent annual growth, and design cycles compressing from months to weeks.
In physical AI terms Fainite represents physics-constrained surrogates under limited data. Two technical claims stand out: independence from large historical datasets, which differs from mainstream surrogates requiring massive simulation corpora and implies physics-informed methods, and physics-enforced modeling, which suggests physical structure encoded into the model rather than merely a soft residual loss.
One framing point matters. This is an early company with a strong team, minimal funding and unvalidated product. The team combines academic depth from Caltech, ETH and Cambridge with industrial experience from Dassault SIMULIA and Google, which is rare among early physical AI startups, yet public funding is 150,000 francs, customer cases are zero and the 10 second versus 48 hour claim is company framing. Key questions are whether low-data methods reach engineering accuracy, whether physics-enforced architecture generalizes across physical domains, and how far 150,000 francs can go.
Company Overview
Team Background
The team is the most notable asset. Donzelli's path runs engineering, industry, finance and founding: a Caltech master's in computational mechanics, simulation and product development at Saipem and Optotune, strategy consulting at UBS, and academic work on automated measurement and quality control for Porsche electric axles. That breadth helps enterprise sales since he speaks both engineering and procurement.
Liu is a Cambridge professor with earlier posts at Caltech and ETH working on computational mechanics and machine learning. Bonvin comes from ETH. Pfeifer's arrival signals large-scale ML engineering capability after twelve years at Google. Keene's arrival signals understanding of industrial software sales with long cycles, high prices and demands for validation and support.
Stage and Funding
The 150,000 franc Venture Kick grant is a Swiss non-dilutive instrument in stages, earmarked for expanding into new engineering domains, accelerating a scalable platform and strengthening team and go-to-market. The stage implies no product market fit evidence yet, with minimal capital testing whether physics-aware AI reaches usable accuracy on real industrial problems, typically against a six to twelve month milestone before a priced round.
Workflow
The disclosed five steps reflect engineering discipline: a free pre-assessment identifying the highest-impact bottlenecks and quantifying expected gains, a technical deep dive analyzing workflows, datasets, solvers and modeling assumptions, model preparation tailoring a physics AI model validated against simulation benchmarks, platform deployment to a secure environment supporting new geometry, boundary conditions and materials, and continuous optimization scaling to more product lines and domains.
The pattern is service-led, since customers rarely pay for an unvalidated tool but will pay for a diagnosis of their bottleneck.
Technology
Why Limited Data Matters
Surrogates follow two paradigms. Data-driven approaches train neural networks on large high-fidelity corpora mapping parameters to fields, with the problem being data cost, since ten thousand cases may need tens of thousands of GPU hours and new products or materials often lack history. Physics-constrained approaches encode conservation laws, constitutive relations and boundary conditions into the model, substituting physics for data, spanning physics-informed neural networks with residual losses, physics-enforced architectures with conservation structure such as equivariant or energy-conserving networks, and operator learning.
Fainite's claim of independence from large datasets places it in the second camp, implying predictions from limited data at the cost of harder training and validity limited to the assumed equations.
Physics-Aware Versus Physics-Enforced
The two terms differ. Physics-aware usually means physics informs training through residual losses without guaranteeing strict compliance at inference, while physics-enforced usually means outputs satisfy constraints structurally such as conservation or symmetry.
The latter is harder but generalizes better since outputs cannot violate physics. Whether Fainite achieves enforcement or merely awareness is undisclosed, and the difference determines whether it has a defensible technical moat or resembles the large PINN literature.
Role of the AI Agent
The agent guides users through multi-step setup and offers expert recommendations from historical cases, the same pattern as PERA AgentX and MATLAB Copilot, encoding expert knowledge into process to lower the barrier.
It presupposes a reliable underlying model. If predictions are inaccurate the agent merely helps users reach wrong answers faster.
Reading the Performance Claim
The 48 hour to 10 second contrast needs care. Forty-eight hours is a full conventional cycle including modeling, meshing, solving and post-processing, while ten seconds is a forward pass. A fair comparison measures total time from geometry and conditions to usable result, and if Fainite requires data preparation, training and validation, the first total may far exceed ten seconds with speed gains only from the second run onward, consistent with all surrogates.
Technical Shortfalls
First, no accuracy validation against commercial solvers including error bounds, applicable conditions or failure boundaries.
Second, unclear physics coverage, whether one model serves thermal, aerodynamic and structural domains or each is trained separately.
Third, the boundary of low-data methods depends on how well equation assumptions match real physics, degrading under strong nonlinearity, multiphysics coupling and material nonlinearity.
Fourth, undisclosed engineering details such as hardware requirements and deployment dependencies.
Product and Commercial Model
Product Form
The platform takes customer CAE data, trains a tailored model, deploys to the customer environment and answers queries, delivering acceleration, simplification and reuse.
Business Model
Inferred from the workflow, the model is pre-assessment possibly free, technical deep dive as consulting, model customization as a project and platform subscription for continuity, the standard industrial AI sequence of proving value before charging. Pricing may combine model count, seats and compute.
Target Customers
Targets are hardware firms with heavy simulation in aerospace, automotive, energy, electronics and materials, all with high simulation cost, many iterations and willingness to pay for efficiency. The 17 million or 9 million engineer figures overstate the addressable market since most engineers do not simulate.
Competitive Position
In physical AI simulation Fainite is earliest stage, earlier than PhysicsX, Vinci and Neural Concept, differentiating on limited data and physics enforcement with disadvantages in capital and customer validation.
Competitive Landscape
Tiering
The field has three tiers: funded companies with products such as PhysicsX at 300 million dollars, Vinci at 250 million, Neural Concept at 100 million and Luminary at 72 million; early companies with clear technical routes including Fainite, Godela and several European deep tech startups; and corporate projects including Ansys SimAI, Siemens PhysicsAI and the NVIDIA PhysicsNeMo ecosystem.
Relationship With Incumbent CAE
Ansys, Siemens, Dassault and COMSOL will integrate AI into existing products rather than cede ground, pressuring startups since customers may prefer AI inside familiar tools. Fainite's opportunity lies in paradigm difference, serving scenarios incumbent tools cannot handle such as early design with scarce data, rather than merely accelerating existing workflows.
Swiss Ecosystem
The ETH system has strong groups at the intersection of computational mechanics and machine learning, Venture Kick provides non-dilutive early capital and Innosuisse funds innovation projects, an environment friendly to deep tech though the domestic market is small and companies must target Europe and global customers from the start.
Financing and Capital Structure
Current State
The only disclosed capital is 150,000 Swiss francs with no priced round, valuation or revenue. Optimistically the company is very early using non-dilutive funds to validate technology while preserving equity, and cautiously it has not yet attracted institutional investment.
Capital Needs
At Zurich research costs of 120,000 to 180,000 francs per senior engineer, the grant covers one or two people for a year, clearly insufficient for six core members including two professors, implying part-time involvement or other sources such as research grants or consulting.
Next Steps
The company must demonstrate usable accuracy against a commercial solver on at least one real industrial case and secure a first paying customer or pilot within six to twelve months, which determines whether a seed round follows.
DISCREPANCIES
1. Two engineer counts. The careers page cites 17 million engineers building physical products while the Venture Kick release cites 9 million hardware engineers affected by simulation bottlenecks. The scopes differ, one being all engineers and the other hardware engineers affected, and this report labels both. Source tier: OFFICIAL (company material) with inconsistent definitions.
2. Framing of the performance comparison. The company cites 48 hours or more conventionally against 10 seconds with Fainite without specifying which steps the 48 hours include or whether the 10 seconds covers training and validation. This report decomposes the comparison in the technology section. Source tier: company marketing requiring technical reading.
3. Meaning of independence from large datasets. The phrase could mean substituting physics for data or fine-tuning a pretrained model on small data, which differ technically. This report adopts the physics substitution reading and notes the method is undisclosed. Source tier: OFFICIAL (statement) plus inference.
4. Framing of the funding stage. Only the 150,000 franc Venture Kick grant is disclosed without stating whether other sources exist, and some interpretations treat it as the only funding. This report labels it the only publicly disclosed capital. Source tier: OFFICIAL (Venture Kick announcement).
5. Market size arithmetic. The company cites a 27 billion dollar market doubling in five years at 10 to 13 percent annual growth, yet doubling implies roughly 15 percent compound growth, inconsistent with the stated range. This report retains the figures and flags the arithmetic. Source tier: company material without cited source.
6. Team size disclosure. Six core members are named including two professors and an advisor without full-time ratios, and professors typically have teaching and research duties. This report does not infer full-time proportions. Source tier: OFFICIAL (member list).
7. Founding date. Some sources say 2025 while registration date is unpublished. This report uses 2025. Source tier: MEDIA crossed with OFFICIAL.
8. Implementation of physics enforcement. The company uses the term physics-enforced without stating whether constraints are architectural such as equivariant networks or loss-based such as PINNs, which differ in difficulty. This report presents both possibilities. Source tier: OFFICIAL (statement) with implementation absent.
Risks and Diligence Gaps
First, minimal funding. At 150,000 francs for six core members the capital is clearly insufficient, and failure to raise a seed round within six to twelve months could stall the project.
Second, missing technical validation, with no public comparison against commercial solvers to judge whether limited data plus physics enforcement reaches engineering accuracy.
Third, zero customers, with no public cases, pilots or paid contracts, in a market where industrial buyers are extremely conservative.
Fourth, uncertain team commitment, since time allocation by two professors and a former Google engineer is unknown and advisory-only involvement would weaken execution.
Fifth, competitive pressure from funded peers already shipping and from corporate AI features, narrowing the window for early entrants.
Sixth, limits of the Swiss market, with few domestic customers and internationalization requiring sales networks and local support unaffordable at this funding level.
Seventh, the applicable boundary of physics enforcement, which depends on equation assumptions and may fail under strong nonlinearity, multiphysics coupling and material nonlinearity, precisely where industrial customers need help.
Composite Assessment
Technology scores 7.0. The limited-data plus physics-enforced route is academically grounded and differentiates from big-data surrogates, and the team combines computational mechanics and machine learning. Deductions for undisclosed implementation, unvalidated accuracy and unclear physics coverage.
Business scores 4.5. The workflow from pre-assessment through deployment fits industrial software sales logic and target customers have real needs. Deductions for zero customers and revenue, no pricing and minimal funding.
Finance scores 4.0. The non-dilutive Venture Kick grant preserves equity. Deductions for a very small amount, no institutional investment and no valuation.
Product scores 5.5. The three capabilities of acceleration, simplification and reuse are directionally right and the AI agent lowers the barrier. Deductions for no public interface, documentation or trial and unverifiable performance data.
Market scores 6.5. Physical AI simulation is hot, customer demand is real and the Swiss deep tech ecosystem is friendly. Deductions for intense competition, conservative buyers and high education cost.
Overall 5.5. This is an early company with a strong team, little capital and no validation: it built technical credibility from Caltech, ETH, Cambridge, Dassault and Google and carved a position distinct from incumbents through limited data and physics enforcement. The distance from 150,000 francs to product market fit is very long. The judgment on physical AI is that physics constraints are the right direction for data scarcity and the capability industrial simulation most needs, while Fainite currently has a thesis and a team rather than verifiable technical and commercial evidence.
References
- Fainite website, accessed 2026: Physics AI Engine positioned to accelerate, simplify and automate simulation workflows with physics-aware deep learning, an AI agent and result reuse; contrast of 48 hours or more conventionally against 10 seconds with Fainite; customer pains of computation above ten hours, more than ten tools and thousands of files
- Team page from the website: Alex Donzelli as CEO with a Caltech master's in applied mechanics and prior roles at Saipem, Optotune and UBS plus academic work on Porsche electric axle measurement and quality control; Professor Burigede Liu as chief scientist from Cambridge with earlier posts at Caltech and ETH; Matthias Bonvin as machine learning lead from ETH; Jan Pfeifer as machine learning scientist after twelve years at Google and core contributions to TensorFlow Lattice, Ranking and Decision Forests; Roger Keene in commercial strategy after Dassault SIMULIA worldwide vice president of sales and operations; Professor Dennis Kochmann as technical advisor from ETH
- Careers page market and mission statements: simulation remains a bottleneck with millions of engineers waiting hours to days; goal of a foundation model for hardware engineering turning simulation into inference; market framing of 17 million engineers, a 27 billion dollar simulation software market doubling in five years at 10 to 13 percent growth and design cycles compressing from months to weeks
- Venture Kick funding announcements from startupticker.ch, immo-invest.ch and midmarketnow: 150,000 Swiss francs, about 170,000 euros, for expanding into new engineering domains, accelerating a scalable platform and strengthening team and go-to-market; technology claims accurate prediction without relying on large historical datasets
- Team and positioning in the same announcements: founders from Caltech, ETH Zurich, Cambridge and Google including Donzelli, Liu and Bonvin plus former executives from leading simulation software companies; targeting 9 million hardware engineers affected by bottlenecks; starting from the most complex engineering problems general-purpose AI cannot solve
- Five-step workflow from the website: free pre-assessment identifying bottlenecks and quantifying expected gains, technical deep dive, model preparation validated against benchmarks, platform deployment to a secure environment, continuous optimization and scaling
- Technical background from public sources: the two surrogate paradigms of data-driven training on large high-fidelity corpora and physics-constrained encoding conservation laws and constitutive relations; physics constraint implementations including physics-informed neural networks with residual losses, physics-enforced architectures with conservation structure such as equivariant or energy-conserving networks, and operator learning
- Technical background from public sources: the difference between physics-aware, where physics informs training without guaranteeing inference compliance, and physics-enforced, where outputs satisfy constraints structurally; the latter generalizes better but is harder to implement, and physics constraints degrade where equation assumptions diverge from real physics under strong nonlinearity, multiphysics coupling and material nonlinearity
- Industry background from public sources: tiering of the physical AI simulation field into funded companies with products such as PhysicsX at 300 million dollars, Vinci at 250 million, Neural Concept at 100 million and Luminary Cloud at 72 million, early startups including Fainite and Godela, and corporate projects including Ansys SimAI, Siemens PhysicsAI and NVIDIA PhysicsNeMo
- Swiss ecosystem from public sources: strong ETH Zurich groups at the intersection of computational mechanics and machine learning, non-dilutive staged Venture Kick funding, Innosuisse innovation grants, and a small domestic market requiring early international focus
- Market data from public sources: third-party estimates of the simulation software market vary between vendor annual reports and research firms, with annual growth typically in the 8 to 12 percent range
- Business model background from public sources: the typical industrial AI sales path of free diagnosis or pre-assessment, consulting, custom projects and platform subscription, because customers are conservative about unvalidated tools and need value proven before paying
This report is compiled from public sources for informational purposes only and is not investment advice. The company is very early with only a 150,000 franc Venture Kick grant disclosed, no customers, revenue or pricing, and performance claims are company statements without third-party verification.
Related papers
- Fourier Neural Operator (Original Paper)
- DeepONet (Original Paper)
- Transolver: A Transformer Solver for Large-Scale Industrial Geometries
- PhysicsNeMo: Neural-Operator-Based Physics Simulation Framework
- Physics-Informed Learning Annual Review 2025-2026
- Sparse-Observation Atmospheric Thermal Forecasting with Physics-Informed Neural Networks for Climate-Aware Digital Twins
- Analysis of trade-offs in urban heat mitigation using a Bayesian Optimization framework for an urban canopy layer model
- Physics-Informed Neural Operator Surrogate for 2D Magnetohydrodynamic Reconnection
Related cases
- DrivAerML High-Fidelity Road-Car Aerodynamics Dataset and Surrogates
- AirfRANS High-Fidelity Airfoil Surrogate Benchmark Case
- Physics-Informed CNN for Battery Pack Temperature Fields
- IASC-ASCE Structural Health Monitoring Benchmark and Physics-Informed Damage Identification
- Wind Turbine Aeroelastic Simulation and Wake Control: A Verifiable Chain from OpenFAST to Reinforcement Learning
- OpenFOAM Industrial Multiphysics Pipeline: From Meshing to Data Supply
- Semiconductor TCAD Surrogates: FP-DDM Domain Decomposition and Etch Process Surrogates
Related reports66
- 51World Deep Dive: A Listed Digital Twin Company Where 51Sim Becomes the First Growth Engine While 51Earth Nearly Vanishes, and Two Irreconcilable Financing Naming Conventions Must Both Be Kept
- AMI Labs: A USD 1.03B Seed Bet on World Models, LeCun's Non-Generative Path and Its Blank Scorecard
- Anshi Pacific (PERA) Deep Dive: A Beijing Domestic-CAE Leader Founded in 2003 Whose PERA SIM General Simulation, AIROM Reduced-Order Models, AgentX Multi-Agent and Jingzhi iGPT/Whale Span from Industrial Simulation to Physics AI
- AspenTech Deep Dive: The Process-Simulation Incumbent Acquired by Emerson and Its Hybrid Industrial-AI Route
- Autodesk Deep Dive: Neural CAD Foundation Models and a $200M World Labs Bet on Physical-World AI, with NAVASTO Adding CFD Surrogate Capability
- Bentley Systems Deep Dive: iTwin Digital Twin and Generative AI Empower Full-Lifecycle Infrastructure Engineering
- BeyondMath Deep Dive: A Foundation Physics AI on the Neural-Operator Route, a $19M Consortium Budget, and Three Misread Facts
- Cadence Deep Dive: System-Level Multiphysics EDA and the Millennium M2000 Supercomputer Anchor Physical AI, with AuraStack Agentic AI and Fidelity GPU CFD as Key Levers
- COMSOL Deep Dive: Four Decades of Multiphysics Solvers, Entering Physical AI via DNN Surrogate-Model Training and All-Physics GPU Acceleration
- Dassault Systèmes Deep Dive: 3DEXPERIENCE and SIMULIA Virtual Twin Physics Behavior Anchor Industrial AI, with LEO Virtual Companion and GPU Explicit Dynamics as Physical-AI Levers
- DP Technology: From the Gordon Bell Prize to National-Level Unicorn Status and the Door of a Hong Kong Listing, the AI-for-Science Leader's Technical Delivery and Commercial Blank
- Emmi AI: From Linz Spin-out to a EUR 330M Mistral Acquisition in 17 Months
- Fysics AI: The PhysX Founder's Second Act, a Domestic Differentiable Physics Engine and Its Ecosystem Play
- Frontop (301313) Deep Dive: From Digital Exhibition Halls to a Physical-AI Data Foundation - FTE Engine, FunSim Simulation and the Gap Between Positioning and Revenue
- Godela Deep Dive: A YC X25 Physics-AI Simulation Startup Whose Ex-Apple, Google and Intel Team Uses Physics-Constrained Surrogate Models to Cut Simulation from Days to Seconds, at Seed Stage with Pilot-Only Access
- HollyWave (688682) Deep Dive: RDSim AI Electromagnetic Simulation and Engineering Agents Close the Simulation-Measurement-Product Loop
- IAT (Alt) Deep Dive: China's First Listed Independent Automotive Design Company Pivots to AI-native R&D with the Yuanfang, Yufeng and Taiyi 3.0 Product System and an NVIDIA Omniverse Partnership, Reporting RMB 975M Revenue and a Widened RMB 250M Net Loss in 2025
- Inductiva Deep Dive: A Closed Case in Open-Solver Orchestration, Eight Premises That Must Be Corrected, and Where Co-Founder and CTO Hugo Penedones Went Next
- Intesim (INTESIM) Deep Dive: A Home-grown Chinese CAE Leader Founded in 2009 with Self-developed Solvers for Structure, Fluid, Electromagnetics, Thermal and Acoustics and Multiphysics Coupling, a ~RMB 100M B Round from CNII and Shenzhen Capital, and 2025 Revenue and Bookings Both above RMB 100M
- iSoftStone (301236) Deep Dive: A ChiNext-Listed IT Services Giant Whose TianShu Platform Builds a Three-Layer Physical-AI Stack with Self-Developed TwinHub Digital Twin, FSim Industrial Simulation, CAE Multiphysics and Phyxis Embodied Control
- Jua Deep Dive: The Earth Physics Transformer as a Large Physics Model for Energy Trading
- Keysight (with ESI Group) Deep Dive: Virtual Prototyping and Hybrid AI Bridge Measurement and Simulation, with Natural-Language Assistants in EDA
- Lightwheel Deep Dive: A Physical AI Data and Evaluation Infrastructure Startup Founded by a Former NVIDIA Simulation Lead, Seven Rounds Totalling Roughly CNY 2.65 Billion at a Valuation Above USD 2 Billion, Seats on Both the Newton and EgoVerse International Standards Committees, With Two Incompatible Accounts of Q1 Orders
- Luminary Cloud: The Physics AI Model Factory Backed by NVentures
- MathWorks Deep Dive: MATLAB/Simulink and Simscape Multiphysics Power Physical AI, with R2026a Copilots and an Agentic Toolkit for Edge Intelligence
- Maniformer (Mifeng) Deep Dive: An Agibot-Spun Physical-AI Data Infrastructure Firm Whose Bodyless MEgo Capture Hardware Is in Mass Production and Which Closed Three Rounds in Half a Year Led by China Telecom
- Mistral AI: How Physics AI Moved From Language Models to EUV Lithography and Nuclear Power
- Monolith AI: The British Engineering-AI Firm That Cut Physical Test Counts Instead of Accelerating Simulation, Now Inside CoreWeave
- Monumo Deep Dive: A Cambridge Deeptech Company Redesigning Electric Motors with the Anser AI Engine, Tens of Millions of Multiphysics Simulations per Day, GBP 10.5M Raised in Total, and a Series A Being Raised
- NAVASTO: Moving Solver Insight Onto the Designer's Desk, and How Autodesk Bought That Entry Point
- Navier: Handing CFD Workflows to Agents, a Seed-Stage Bet on AI for Hardware Engineering
- Neural Concept: An EPFL Spin-out Betting on Neural Surrogates, with $138M Raised
- NP Co. (NP Company) Deep Dive: A 2025-founded Paris Inria-born AI-native simulation startup that raised a EUR 6M pre-seed led by Partech with Mistral AI co-founders as angels, claiming a pre-trained physics foundation model at 1,000x speedup — most claims unverified
- NVIDIA Deep Dive: Building the Full-Stack Physical-AI Infrastructure with PhysicsNeMo, Warp, Cosmos World Foundation Models and Omniverse
- Physical Intelligence Deep Dive: A $2.1B Bet on the GPT Moment for Robotics, With pi0.7 Self-Reporting 60-80 Percent Success on Unseen Tasks and Neither Pricing nor Revenue Disclosed for the Pi API
- Physical Superintelligence Deep Dive: A $58M Seed Bet on Virtual Physicists, With Emmy Targeting AI Data Center Multiphysics Optimization as Its First Commercial Proof Point
- PhysicsX: Pretrained Geometry and Physics for Industrial Engineering at a $2.4B Mark
- Prometheus: A $41B Mark, $18.2B Raised, and Not Yet a Product
- PTC Deep Dive: Creo/Onshape and the ThingWorx Digital Thread Power Physical AI, with NVIDIA Isaac Sim Bridging Robotics Design to Simulation
- Qianjue Technology Deep Dive: How a Body-Agnostic Embodied Brain Pairing the Predictive World Model Polibrain with Brain-Inspired Partitioning and Polynomial Representation Reached a Hundred Thousand Cleaning and Service Robots
- Quanscient Deep Dive: Cloud-Native and Quantum-Powered Multiphysics Simulation as the Data Engine for AI-Native Hardware Engineering
- Rescale Deep Dive: An HPC Operating System and Desktop Engineering Workbench in the Cloud, a $115M Series D, and Five Misread Facts
- Schrödinger Deep Dive: Physics-Based FEP+ Simulation and Machine Learning Power Molecular Discovery, with the 2026 Bunsen Agentic Co-Scientist
- Secondmind Deep Dive: A Cambridge Engineering-AI Firm Whose Bayesian Active Learning Cuts CAE Simulation and Test-Bench Runs by up to 90%, with Roughly USD 73M Raised and Mazda Both Customer and Investor
- Shichuang Technology Deep Dive: SupreCAST Cloud-Native Material-Processing Simulation and Industrial Agents Advance the CAE+AI Strategy
- ShonCloud (Shunyun) Deep Dive: Meshless Particle-Method CFD for EV and Nuclear Fluid Multiphysics
- Shuzhi Maijin Deep Dive: A Shenzhen Fully Self-Developed SNN-plus-LBM CFD Startup Claiming Second-Level Results on Hundreds-of-Millions of Grids on Consumer Hardware, With a Tens-of-Millions-RMB Angel Round From a Ministry-of-Finance-Affiliated Fund and an FPGA Migration Plan
- Siemens (incl. Altair) Deep Dive: The 2025 Altair Acquisition Unites Simcenter and Altair, with PhysicsAI Geometric Deep Learning and PhysicsAI Generate as the Physical-AI Core
- Silurian Deep Dive: A Generative Forecasting Transformer Earth-System Foundation Model for Weather Risk on Energy and Infrastructure Assets
- SimScale Deep Dive: A Cloud CAE Platform Built on Core-Hour Subscriptions, EUR 52M of Equity Kept Separate from EUR 25M of Debt, and Five Misread Facts
- Skild AI Deep Dive: A General Robot Foundation Model Incubated at CMU's Robotics Institute, With a USD 1.4B Series C at a USD 14B Valuation, a Conflicting Series B Size, and an ARR Figure That Is a Run-Rate Rather Than Revenue
- Songying Deep Dive: A Huawei-Cloud-Alumni Startup Building a Domestic Physical AI Operating System, With ORCA OS Positioning Against NVIDIA Omniverse, Three Incompatible Financing Narratives, and a Verifiable Third-Party Anchor in a CNY 20 Million Minority Investment
- Sudo Tech Deep Dive: Zero-Real-Data Simulation Training, the Fully In-House Sudo R1 Stack, and an Atomic-Skill Commercialisation Strategy from a Startup Valued Above CNY 20 Billion in 17 Months
- Suochen Technology: The Pain of Becoming a Physics-AI Platform from a CAE Vendor
- Synopsys (Ansys) Deep Dive: The 2025 Ansys Acquisition Turns the EDA Leader into a Silicon-to-Systems Engineering Powerhouse, with Ansys SimAI and the NVIDIA Physical-AI Partnership as the Core Physics-AI Levers
- TenFong Deep Dive: An Academician-Led Chinese Industrial Simulation Software Vendor Whose Official Timeline and Database Records Use Incompatible Round Names, With an Official Financing PDF and an Investment Database Both on File
- Tianfu (Nanjing Tianfu Software) Deep Dive: A 2011-founded Chinese Intelligent-CAE Veteran That Closed a Several-Hundred-Million-RMB Series C, Pivoted to an Industrial World Model in 2024, With Unreconcilable Financing and Valuation Records and No Disclosed Revenue
- Tongyuan Soft Control Deep Dive: MWORKS (Syslab/Sysplorer) as a Domestic MATLAB/Simulink Alternative, with MWORKS AI and the MOGICK Engineering Agent Advancing Home-Grown Physical AI
- Vinci AI: Turning Physics Simulation from a Checkpoint into a Continuous Constraint
- Westlake Robotics Deep Dive: Westlake University's First Robotics Spin-out Builds a Dual-Pretrain Brain-and-Cerebellum Architecture with WR1 Whole-Body Foundation Model and GAE Behaviour Expert
- World Labs: Fei-Fei Li's Spatial-Intelligence Unicorn, Acquired by AMD for $8.2B
- XVERSE Deep Dive: A Simulation-World Route to Physical AI Built on the XVERSE Model Family, an Edge-Cloud 3D Engine and the Physics-Based MotionGen Model
- SimWE: From In-House Solvers to a Physical-AI Engine, the Domestic CAE Leader and What Comes Next
- Yunjing Zhifang Deep Dive: A Peking-University-Incubated Cloud-Native GPU CAE Company with the Simversus Platform and the YJ-DORGP Small-Sample Intelligent Solver, Reporting 275 Percent Order Growth, 160 Percent Revenue Growth and Breakeven in 2025
- Zhongke Diwiji (Five Ages) Deep Dive: A CAS-Institute-of-Automation and Tsinghua Spin-out Whose FAM Embodied Foundation Model Uses Action Optical Flow for Cross-Embodiment Generalisation, Reaching 97% Success from 3 to 5 Demos, with EraBrain and Jiling Robots Serving Sinopec, Leapmotor and Midea
- ZWSOFT (688083) Deep Dive: Overdrive Kernel and ZW3D Wukong 2027 Underpin an Integrated CAD/CAE/CAM Stack, with a CAx Agent Platform as the Practical Route to Engineering AI
This report is compiled from public sources. It is for reference only and does not constitute investment advice. Data is current as of the publication date; please refer to the company’s official filings.