Home›Companies›Dualizer
Dualizer

AI-powered physical simulation to design/optimize chemical processes

Dualizer builds AI-powered, physics-based process simulation and digital twins to design and scale chemical processes and optimize existing plants. We help engineering teams identify bottlenecks, increase yield and throughput, and reduce energy and production costs—improving plant efficiency and profitability. The platform combines AI reasoning, our proprietary deterministic simulation and optimization engine, and physics-informed machine learning. Engineers can build, calibrate, and refine models through conversation, starting from a diagram, available data, or an engineering objective. Calculations account for thermodynamics, material and energy balances, equipment behavior, and operating constraints, enabling teams to optimize interconnected equipment and operating conditions across a plant. Equipment sizing and costing, 3D layout planning, and techno-economic analysis connect process improvements to plant-level decisions. Teams can accelerate new process scale-up, debottleneck operating plants, and continually refine operating strategies as production targets and plant conditions change. Dualizer makes rigorous simulation and optimization an ongoing capability for improving industrial productivity and profitability throughout a plant’s life.
Active Founders
Nathan Liu
Nathan Liu
Founder/CEO
UC Berkeley ChemE/CS dropout (’27); Mesbah/Yang Lab: AI/ML researcher, JACS co-author, Kavli ENSI Fellow. Founded Emissions Eye: 17K+ global deployments, 14% NOx reduction (2,100+ tons). Paradromics: 200+ bps preclinical neural decoding. Yzdelta: RTO, >90% thermal efficiency, >99% VOC destruction. CalSol: 5th in US + Fastest Lap. Falling Walls SF Bay Area: 2nd place. VPC Fellow: 1/16 in US. Harvard SVMP; UN Millennium Fellow/Director; With Dualizer: 1st/531 teams in US, $100K Texas A&M AIVVC.
Chris Donahue
Chris Donahue
Founder/CTO
UC Berkeley ChemE dropout (’27); ex-AVEVA; 4+ years experience with rigorous process modeling technologies by AspenTech and AVEVA. Taught simulation certification courses for AVEVA around the USA to industry-professional engineers as an undergraduate. 2.5+ years in Morales-Guio & Christofides Labs: dynamic modeling, optimization and published process-simulation research. Built a rigorous SMR model for UCLA. With Dualizer: 1st/531 teams in US, $100K Texas A&M AIVVC.
Dongha Kim
Dongha Kim
Founder/CSO
UC Berkeley ChemE/EECS dropout ('27); ex-Logitech software engineer. Battery-degradation PINN: 42% lower MAE than XGBoost. LBNL Helms Group: CNN-based ML for semi-automated XPS analysis of battery electrolytes. Materials Project engineer—platform serving 620K+ users, 200M+ annual queries. ML work spanning Nobel laureate Omar Yaghi’s MOF/COF research and QuantumScape solid-state batteries. With Dualizer: 1st/531 teams in US, $100K prize at AIVVC. I'm a benzene(C6H6) enthusiast.
Company Launches
Dualizer — Agentic physics simulations that debottleneck manufacturing
See original launch post

Hi everyone—we’re David, Chris, and Nathan, the founders of Dualizer.

We’re building agentic physics simulation to design, scale, and optimize manufacturing—from everyday essentials like food, beverages, and medicine to the fuels and advanced materials powering aviation and modern computing.

https://youtu.be/W71Gu1w0PwU

Engineering decisions shape a plant’s economics for decades. Yet building reliable models and evaluating improvements remains slow, manual, and specialized.

High production costs can put essentials like food, medicine, and energy out of reach. Inefficient production can waste resources and pollute the environment for decades. Design or operating failures can cause deadly incidents.

Dualizer is an engineering software that automates process modeling, simulation, and optimization—helping engineers design better plants and improve existing ones by balancing yield, throughput, energy use, and cost within real operating constraints. Engineers can build physics-based digital twins from a process diagram, available data, or an engineering objective, then use them to:

  • Design and scale new production: Evaluate equipment sizes, operating conditions, and constraints before committing capital.
  • Improve existing plants: Identify bottlenecks and test changes to increase yield and throughput while reducing energy and production costs.
  • Connect engineering to economics: Compare alternatives through equipment sizing and costing, 3D layout planning, and techno-economic analysis.

Dualizer’s proprietary simulation and optimization engine accounts for thermodynamics, material and energy balances, equipment behavior, and operating constraints. Physics-informed machine learning uses plant data to refine these models, so simulations more closely reflect how each plant actually operates.

Our ambition is to become the engineering foundation for the world’s process industries—helping startups and global manufacturers alike commercialize breakthroughs faster and make essential goods cheaper and more abundant.

Same school. Same year. Same major. We met at UC Berkeley, Class of ‘27, Chemical Engineering. Our research and industry experience in process modeling, physics-based simulation, and AI/ML gave us the foundation for a shared ambition: transforming how industrial plants are designed and operated. We were all headed toward PhDs. We chose to build Dualizer.

Our ask: Know a company that makes chemicals, food and beverages, fuels, medicines, or advanced materials? We’re looking for teams to pilot Dualizer to help scale up production, increase output, or reduce energy and material costs. We’d appreciate introductions to the engineers, plant managers, or founders responsible for how those products are made.

Learn more: dualizer.com

YC Photos
Hear from the founders

How did your company get started? (i.e., How did the founders meet? How did you come up with the idea? How did you decide to be a founder?)

All three co-founders are same school (UC Berkeley), same class (’27), same major (Chemical Engineering). Nathan and David have been classmates and friends since their first year of college and later met Chris.

The idea was simple: designing and improving chemical processes remains far too expensive, difficult, and time-consuming. We were all on paths toward PhDs, but chose to tackle this persistent industry problem as founders. We started Dualizer to bring our education and research and industry experience in process modeling, physics-based simulation, machine learning, and AI for science directly into engineering practice.

What is the core problem you are solving? Why is this a big problem? What made you decide to work on it?

Manufacturing spans an estimated 7.5 million factories worldwide across all sectors. Process engineering underpins the production of fuels, medicines, food, cosmetics, household products, fertilizers, plastics, textiles, packaging, and advanced materials—shaping the cost and availability of everyday goods.

A large plant can require billions of dollars to build and operate for decades. Poor design and inefficient operation trap capital in underperforming facilities, constrain production, waste resources, and turn promising technologies into failed investments. The consequences extend across supply chains: higher prices, shortages, delayed commercialization, and unnecessary emissions.

Yet building reliable models, calibrating them against plant data, and evaluating improvements remains slow, manual, and specialized. Limited engineering capacity leaves valuable improvements unexplored and new processes expensive to scale. We encountered that constraint through our own research and simulation work and started Dualizer to make rigorous process design and optimization accessible throughout a plant’s life.

What is your long-term vision? If you truly succeed, what will be different about the world?

Our ambition is to make Dualizer the engineering foundation for the world’s process industries—enabling startups and global manufacturers alike to accelerate process development, scale production, and maximize plant performance.

If we succeed, industry will produce more from existing assets, commercialize breakthroughs faster, and make essential goods cheaper and more abundant—with less capital, energy, and material required per unit of output.

In practice, that could mean:

  • Greater energy security: Increasing throughput and reducing costs across gas processing, LNG, refining, and fuel production.
  • More productive industrial capital: Unlocking capacity in chemical, fertilizer, cement, and materials plants while reducing the need for costly expansions.
  • Faster AI and electrification growth: Scaling mineral refining, battery materials, and the high-purity chemicals and gases needed for semiconductor manufacturing.
  • Profitable industrial decarbonization: Making lower-carbon fuels, hydrogen, cement, and carbon capture competitive at commercial scale.
Jobs at Dualizer
US / Remote (US)
$125K - $175K
0.10% - 1.00%
3+ years
US / Remote (US)
$125K - $175K
0.10% - 1.00%
3+ years
US / Remote (US)
$200K - $400K
0.50% - 5.00%
3+ years
Dualizer
Founded:2026
Batch:Fall 2026
Team Size:3
Status:
Active
Location:San Francisco
Primary Partner:Christopher Golda