Tech Soft 3D, a U.S. developer of engineering software toolkits for visualization and data exchange, has launched HOOPS AI, a framework designed to make computer-aided design (CAD) data accessible for machine-learning workflows. The system integrates CAD access, dataset preparation, and encoding in a single environment, enabling data scientists to move directly from raw geometry to reproducible models without relying on external conversion tools.
Machine learning has reshaped manufacturing and product development, yet CAD information—central to engineering design—remains difficult to use in algorithmic training. Converting 3D geometries into machine-readable formats usually requires fragmented scripts, manual corrections, and dependence on expensive proprietary systems.The software eliminates these barriers by combining access, preparation, and encoding in one reproducible pipeline. Built on HOOPS Exchange, Tech Soft 3D’s established CAD import and export library, the framework offers direct, high-fidelity access to over thirty file formats covering geometry, topology, assemblies, product manufacturing information (PMI), and metadata through a Python API. This architecture avoids risky intermediate conversions and reduces reliance on native CAD applications.
According to Gavin Bridgemann, Chief Technology Officer at the firm, “HOOPS AI represents a major leap forward for organizations looking to finally harness artificial intelligence for 3D CAD. It provides a complete, reproducible pipeline that makes machine-learning workflows with CAD data both practical and scalable.” The toolkit manages ingestion, versioning, and experiment tracking for traceable results while including utilities for visualization, segmentation, cleaning, and encoding so teams can generate consistent datasets for model development.
Tech Soft 3D says the system automates large-scale dataset preparation and end-to-end workflow orchestration across the full machine-learning lifecycle. Storage and logging components record every experiment to ensure repeatability and traceability—capabilities required for verifiable research and model validation. Visualization tools embedded in the environment allow inspection of intermediate stages, giving users insight into dataset integrity before model training begins. By integrating these stages into one controlled pipeline, HOOPS AI eliminates the fragile, script-based workflows that often slow data-driven engineering.
The company’s established role in the CAD-software ecosystem underpins its ability to support this new platform. Its technology is embedded in tools from Ansys, Siemens, Hexagon, NVIDIA Omniverse, Unreal Engine, and Unity 3D, providing the data translation backbone for many industrial and visualization applications. Experience processing large volumes of CAD data for these vendors forms the basis for HOOPS AI’s focus on reliability and scale. With more than 750 independent software vendors already using its software development kits, Tech Soft 3D plans to extend HOOPS AI access to this network and to research groups developing AI methods for design, simulation, and manufacturing.
Attempts to apply machine learning to CAD data have been constrained by inconsistent formats and limited interoperability between platforms. Academic research has explored using neural networks for generative design and feature detection, but translating industrial CAD models into usable datasets remains costly and time-consuming. In commercial software, progress has centered on proprietary solutions; Siemens has integrated AI-assisted design within its Xcelerator platform, while Autodesk’s research division has examined automation of repetitive modeling tasks. These examples illustrate the absence of a universal, vendor-neutral framework capable of connecting CAD geometry with machine learning—an issue HOOPS AI directly addresses.
Founded in 1996 and headquartered in Bend, Oregon, Tech Soft 3D operates offices in France, England, Japan, Germany, and Norway. Backed by investment firm Battery Ventures, the company provides software development kits used for CAD and CAE data conversion, visualization, and simulation across engineering and manufacturing sectors.
Tech Soft 3D has opened a technology preview to allow developers and partners to evaluate performance within existing workflows. Interested users can request access through the company’s website at techsoft3d.com/developers/products/hoops-ai.
Help shape the 2025 3D Printing Industry Awards. Sign up for the 3DPI Expert Committee today.
Ready to discover who won the 2024 3D Printing Industry Awards?
Subscribe to the 3D Printing Industry newsletter to stay updated with the latest news and insights.
Featured image shows 3D geometry processed by HOOPS AI framework. Image via Tech Soft 3D.
