From knowledge to new materials.
We’re building an AI-powered platform to discover new materials by connecting scientific knowledge, predictive models, and physical testing.
Our aim is to draw on the accumulated knowledge of materials science, predict promising new candidates, and test them in the physical world—with each research cycle contributing evidence to guide further discovery.
Discuss a materials challenge
02 / Material possibilities
New materials.
New capabilities.
The properties of materials shape what we can build. We’re developing a platform to explore new compositions and structures against the requirements that matter in real applications.
01Can we discover stronger, lighter materials?
Explore how changes in composition and structure could meet mechanical requirements at a lower weight. The starting point is the application: what the material must support, withstand, and retain over time.
Discuss this materials challenge02Can materials perform reliably in more demanding environments?
Temperature, corrosion, pressure, and repeated loading can limit material performance. These conditions help define the properties to investigate and the evidence needed to evaluate new candidates.
Discuss this materials challenge03Can essential performance require fewer resources?
Investigate alternative compositions and processing routes that could reduce material use, energy requirements, or dependence on scarce inputs while meeting the needs of the application.
Discuss this materials challenge03 / How it works
Research. Predict.
Test. Learn.
Our platform is being designed to connect materials research, AI prediction, physical testing, and model training in a continuous discovery loop.
- 01
Connect existing knowledge.
Bring together scientific literature, materials data, and prior experimental findings to identify relevant patterns and frame the discovery challenge.
- 02
Predict new material candidates.
Use AI to propose material compositions and structures, predict relevant properties, and prioritize candidates for investigation.
- 03
Test candidates. Train the engine.
Evaluate candidates through physical testing, then feed the results into the platform’s knowledge base and training process to refine future predictions.
Each experiment adds evidence for the next prediction.
The learning loop
Test the prediction.
Advance the search.
An experiment can support a prediction, reveal its limits, or point in a new direction. We’re building an engine designed to carry that evidence into future research.
04 / About Thetic
New materials.
Greater possibilities.
The materials available to us shape what we can build. Thetic’s ambition is to expand those possibilities by developing a repeatable, AI-powered approach to discovering useful new materials.
Our long-term goal is a discovery engine that carries knowledge from one investigation into the next—expanding its evidence base, refining its models, and building the capacity to pursue increasingly ambitious materials challenges.
05 / Get in touch
Help shape the future of materials.
We welcome conversations with companies facing materials challenges, potential research collaborators, and investors interested in the future of discovery.