Inviscid AI
Product & Competitive Intelligence
Creates digital twins of buildings and data centers to autonomously optimize energy and cooling.
Company Overview
Builds a real-time, physics-informed AI simulation platform that creates digital twins of commercial buildings and data centers to autonomously optimize energy consumption and operational costs using Physics-Informed Neural Networks (PINNs).
Competitive Advantage & Moat
Product Roadmap & Public Announcements
Inviscid AI describes real-time digital twin simulations for commercial buildings and data centers, neural network surrogates replacing traditional CFD solvers, and direct BMS/IoT integration for closed-loop HVAC and energy optimization. Autonomous operational optimization with physically plausible AI recommendations grounded in thermodynamic and fluid dynamic laws.
Signals & Private Analysis
Two-person YC-backed team in deep R&D and early customer discovery. Co-founder Ziming Qiu's NYU ECE doctoral research in computer vision and ML suggests ongoing academic collaboration. Industry signals point toward pilot programs with data center operators. Digital twin architecture positions them for expansion into climate resilience modeling, predictive maintenance, and grid-interactive building controls.
Product Roadmap Priorities
Real-time HVAC airflow simulation using Physics-Informed Neural Networks to replace weeks-long CFD analysis with seconds-fast digital twin inference for commercial buildings.
Instead of waiting weeks for engineers to simulate how air moves through a building, Inviscid AI's neural networks do it in seconds and automatically adjust the HVAC system to save energy.
It's like replacing a weather forecast that takes a week to compute with a meteorologist who can predict the exact breeze in every room of your office building before you even feel warm.
Predictive digital twin for data center thermal management that autonomously optimizes cooling infrastructure to prevent hotspots and reduce power usage effectiveness (PUE).
Inviscid AI builds a virtual replica of your data center that predicts where heat will build up before it happens and automatically adjusts cooling to prevent outages while slashing electricity bills.
It's like having a chess grandmaster who can see 50 moves ahead playing against your data center's heat — except every move also lowers your electric bill.
Occupancy-adaptive energy forecasting and demand response optimization that dynamically adjusts building systems based on predicted occupancy patterns and grid signals.
Inviscid AI predicts how many people will be in each part of a building throughout the day and automatically pre-adjusts energy systems to avoid expensive peak electricity charges — and even gets paid by the utility for doing it.
It's like a restaurant chef who knows exactly how many guests are coming, pre-preps everything during off-peak grocery prices, and then gets paid by the power company for not turning on the deep fryer during rush hour.
Company Overview
Key Team Members
- Ziming Qiu, Co-Founder
- Kabir Jain, Co-Founder
Founder Force Multiplier
Inviscid AI combines deep academic expertise in Physics-Informed Neural Networks with real-time digital twin architecture, enabling simulations that are orders of magnitude faster than traditional CFD while remaining physically accurate, a capability most competitors lack entirely.
Funding History
- 2025 | Kabir Jain and Ziming Qiu co-found Inviscid AI.
- 2025 | Accepted into Y Combinator batch.
- 2026 | No public funding rounds disclosed; estimated ~$500K (YC standard deal).
Competitors
- Digital Twin Platforms: Willow (building digital twins), Autodesk Tandem.
- AI Building Optimization: BrainBox AI (autonomous HVAC), PassiveLogic (autonomous building control), Turntide Technologies.
- Traditional BMS/Analytics: Siemens Building X, Honeywell Forge, Johnson Controls OpenBlue, SkySpark.
- Data Center Cooling: Nautilus Data Technologies, Colovore, Schneider Electric EcoStruxure.
