BEND, OR, Mar 31, 2026 – Tech Soft 3D has released HOOPS AI following a beta program with more than 30 companies, targeting teams that need to bring complex 3D engineering data into machine learning workflows. HOOPS AI supports software vendors and engineers across PLM, MES, CAM, manufacturing, and other data-intensive 3D workflows.
Machine learning has advanced across industries, but CAD data remains difficult to use in these workflows. Current approaches rely on fragile processes and custom scripts that increase cost and development time. HOOPS AI addresses this gap with a framework for building machine learning models using CAD data.
“The official launch of HOOPS AI marks an important step in Tech Soft 3D leading the effort to bring AI to engineering data,” said Gavin Bridgeman, CTO, Tech Soft 3D. “For the first time, developers have a framework designed specifically to make CAD data usable in modern machine learning workflows.”
HOOPS AI focuses on preparing CAD data and testing models at scale. It allows teams to run hundreds of models in parallel, enabling iteration across model variations. The platform supports use cases such as part classification, metadata enrichment, feature detection, similarity search, duplicate detection, and design reuse.
HOOPS AI builds on Tech Soft 3D’s CAD translation technology, used by partners including Ansys, Siemens, Hexagon, NVIDIA Omniverse, Unreal Engine, and Unity. The technology reflects decades of CAD data processing and supports integration of machine learning workflows with engineering data.
New Features for the Full Release of HOOPS AI
HOOPS AI now supports Windows and Linux, enabling use in AI and data processing pipelines and integration into machine learning infrastructure.
The release also introduces CAD embeddings, a technology that capture relationships within complex CAD data. This allows machine learning models to identify similar parts and analyze design context, with relationships derived from data patterns rather than human input.
Machine Learning Workflow Support
HOOPS AI supports every stage of the machine learning lifecycle, including CAD data access, dataset preparation, encoding, visualization, and scaling. It allows teams to run hundreds of model variations, track outcomes, and refine models through repeated testing.
HOOPS AI applies machine learning to CAD data for tasks such as part classification, feature detection, and similarity search across design libraries. It also adds Python access to expose additional CAD data, including product manufacturing information (PMI). Current models are trained on public datasets, with support for training on private data.
Source: Tech Soft 3D
About Tech Soft 3D
Tech Soft 3D, founded in 1996, provides engineering software development toolkits (SDKs) and industrial applications for 3D data conversion and visualization. Headquartered in Bend, OR, with additional offices in France, England, Japan, Germany, and Norway, the company serves industries such as CAD, CAE, and CAM. Their product offerings include solutions for CAD data translation, visualization, and data publishing, facilitating the development of applications across desktop, mobile, and web platforms. Tech Soft 3D employed around 250 individuals.