ok_computer2.0 / Innovation in motion
The application series / 01—03Accelerate the
next breakthrough.
Bold ideas deserve a path
from possibility to proof.
Cutting-edge scientific AI for ambitious materials and engineering research—from new candidates to the next experiment.
Explore the filmsMaterials Energy Infrastructure
Powered by ok_computer2.0Resilient infrastructure
Rethink what a surface can do.
Explore a new generation of paving—connecting recycled polymers, structural performance, and stormwater management in one design process.
Scientific models. Connected decisions. New possibilities.
The innovation question
How can polymer composition, grid geometry and installation work together to carry loads and manage stormwater?
Designed to accelerate discovery
Connect formulation, geometry, and site conditions early, so the next physical test can address a more useful design question.
Explore a partnershipExplore the scienceFrom the first input to the next experiment
01 / Inputs
Define the opportunity.
- Polymer blend and manufacturing history
- Grid geometry, aggregate and pavement layers
- Rainfall, initial moisture, sediment and drainage
- Load history, temperature and installation conditions
02 / Model path
Connect the intelligence.
- CETTE-MIXPROC
Connect polymer ingredients and proportions to an ordered manufacturing route.
- CETTE-MÉSOLIEN
Connect processing and material structure to effective mechanical and transport parameters.
- CETTE-CHAMPOP
Map geometry, material parameters and boundary conditions to hydraulic and mechanical fields.
03 / Output quantities
Focus the investigation.
- Ponding, stored water and outflow over time
- Water balance across the pavement assembly
- Load response, displacement and permanent deformation
- Uncertainty for the stated material and operating conditions
04 / Verification
Compare hydraulic predictions with infiltration and drainage measurements, and mechanical predictions with repeated-load tests on the specified pavement assembly.
Inside the engine
Specialized intelligence.
Shared ambition.
18 architecture components connect scientific questions across scales. Seventeen neural model families and variants, plus a dedicated quantum workflow—each with a defined role in discovery.
Explore all 18 architecture components
| Component | What it does |
|---|---|
| CETTE-ÉQUIAT | Conservative equivariant atomistic learning ÉQUIAT learns how the local environments of atoms contribute to the energy of a material. |
| CETTE-ÉQUIAT-LR | Atomistics with explicit long-range interactions ÉQUIAT-LR extends the atomistic family to problems where interactions beyond a local geometric neighborhood matter. |
| CETTE-ÉQUIAT-MF | Method-conditioned atomistic predictions ÉQUIAT-MF connects calculations made under different, explicitly identified scientific methods. |
| CETTE-INVAR | Invariant property learning and baseline inference INVAR retains a direct route from a material representation to a defined property. |
| CETTE-COMPACT | Compression and efficient deployment COMPACT carries trained scientific capabilities into a smaller deployment model. |
| CETTE-SYMGÈNE | Conditional crystal generation SYMGÈNE explores compositions and periodic structures conditioned on useful properties and permitted chemistry. |
| CETTE-MIXPROC | Mixture and processing design MIXPROC searches the joint space of ingredients, proportions and manufacturing history. |
| CETTE-SYNTHÉVAL | Condition-aware synthesis feasibility SYNTHÉVAL evaluates how a proposed material could be prepared within a real precursor and equipment envelope. |
| CETTE-MÉSOLIEN | Microstructure and scale transfer MÉSOLIEN connects materials chemistry and preparation history to the microstructures that govern engineering response. |
| CETTE-CHAMPOP | Field prediction with neural operators CHAMPOP learns the map from a physical problem specification to spatial or time-dependent fields. |
| CETTE-CINÉTA | Constrained reaction and process dynamics CINÉTA combines learned kinetic information with explicit chemical and engineering structure. |
| CETTE-SUPRAC | Superconducting-material screening SUPRAC organizes superconductivity discovery around composition, structure, phase and measurement conditions. |
| CETTE-FLUXCIR | Superconducting-circuit design and response models FLUXCIR connects Coltrane circuit topology, device geometry and fabrication context to superconducting-circuit behavior. |
| CETTE-SYNDÉCODE | Temporal quantum-error decoding SYNDÉCODE uses the history of quantum-error syndromes to infer the information needed for decoding. |
| CETTE-ÉLECTROFLUX | Fuel-cell electrochemistry, transport and degradation ÉLECTROFLUX links fuel-cell materials to operation at the cell scale. |
| CETTE-ONDEQ | Variational neural electronic wavefunctions ONDEQ represents an electronic wavefunction with a neural model and optimizes it through variational Monte Carlo for a specified Hamiltonian. |
| CETTE-EMBEDQQuantum workflow | Quantum embedding and variational eigensolver research EMBEDQ places a selected electronic subproblem inside an explicit embedding environment. |
| CETTE-NEUROCIR | Neural accelerator and hardware/software co-design NEUROCIR connects the structure of an AI workload to accelerator architecture, software mapping and physical implementation. |
Build with Cette AI
What will you
make possible?
Bring your next research question.
Let’s explore where it could lead.