What does engineering look like when AI works as a collaborator rather than a tool? Over 100 engineers, researchers, and industry professionals explored that question at the CADFEM AI Conference Singapore 2026 at the JW Marriott Singapore South Beach on 26 May. PI Probaligence joined to discuss practical AI in engineering, from simulation to predictive intelligence.
AI in engineering, beyond the buzz
The sessions moved past the AI hype and focused on what already changes engineering work: AI-enhanced structural, thermal, electromagnetic, and multiphysics simulation, predictive engineering, and smart digital applications. The full program is on the official event page.
The audience matched that breadth, with attendees from electronics and semiconductor, aerospace and defence, automotive, and academia. In line with Singapore’s national AI initiatives, one theme ran through the day: real adoption happens when industry, academia, and technology partners work together.
Predictive engineering on the agenda
A highlight was the session on AI-powered predictive engineering with STOCHOS, presented by CADFEM. The response confirmed a clear shift from traditional simulation toward predictive, data-driven engineering. Concrete use cases drew the strongest engagement: practical AI on specific engineering problems sparked more questions than any abstract AI topic.
After the CADFEM AI Conference Singapore
The conference also deepened the collaboration with CADFEM APAC and brought new perspectives from the Southeast Asian market back home. STOCHOS Flow drew particular interest as the workflow layer that puts these methods into daily engineering practice. Thanks to the CADFEM team for the organization and the hospitality. We look forward to building on this momentum across the region.
About PI Probaligence
PI Probaligence GmbH develops local, probabilistic AI for engineering and R&D. Based in Grafing bei Munich and part of the CADFEM Group, PI is an Ansys Technology Partner. STOCHOS and STOCHOS Flow build fast, uncertainty-aware surrogate models from simulation, experimental, and process data. As a result, teams shorten development cycles and make more reliable decisions.