Dan DeDecker’s Post

ZW3D 2025 introduces three innovative AI-powered features: · Command Advisor, Model Match, and API Assistant. · Experience CAD software that understands your needs, predicts your next steps, and streamlines your workflow. Discover how ZW3D 2025 enhances your design process！ #AI #ZW3D2025

I am super inspired daily to work with our brands that are developing first class software for manufacturing and now AI solutions to further enhance the experience. SMF security approved 🤠✔️ Read more about our new AI solution(s) in the article and admire the picture of your fellow security nerd.

Xtend AI is building some amazing tech with OpenBOM & Onshape by PTC. What a pleasure it is working with David Rubli and the #xtendai team. Please take 3 minutes to learn more about them (and us). #openbom

Seeing lots of great founders disrupting EDA with AI Most EDA tools don’t do real “engineering,” but rather take the rote tasks out of implementing engineering output from centuries of “experience” Can AI collapse the heuristics of the best engineers that ever lived to actually automate engineering? Hard to beat EDA at its own game by making it better with AI Generating the boundary conditions that engineers normally use to start the engineering process and better optimized at the end would be a great complement to existing EDA

Which industry do you believe will experience the greatest impact from Generative AI, and why? Generative AI is set to revolutionize the engineering and construction industry by optimizing designs, automating workflows, and enhancing decision-making. Tools like E3D software, commonly used for 3D modeling in engineering, benefit significantly from AI by generating efficient designs, identifying errors early, and offering multiple optimized solutions tailored to project requirements. Machine Learning (ML) in E3D software improves accuracy and efficiency by automating tasks like clash detection, component placement, and error identification. It learns from past projects to provide smart suggestions and predictive analytics, enabling engineers to make better decisions quickly. This reduces manual effort, improves productivity, and minimizes costs. The integration of Generative AI and ML in engineering software allows for faster, more sustainable, and cost-effective project execution. These technologies empower engineers to tackle complex problems, design innovative structures, and ensure high-quality outcomes, making them indispensable in modern engineering practices  #H2STC #GenAI #Hack2skill #H2Sgiveaway Hack2skill

Many experts believe that machine learning will not add significant value to CAE software, but what about CAD? Imagine this: you describe your design concept in plain English, and an AI whirs to life, spitting out a parametric CAD model ready for manufacturing. Sounds like science fiction, right? But generative AI is making waves, and its potential impact on CAD is impossible to ignore. While true text-to-CAD might be a ways off, the ability to use AI to automate repetitive tasks, optimize designs, and even generate creative concepts is closer than you think. What do you think? Will AI be a game-changer for CAD? Or is the human touch here to stay? #CAD #AI #GenCAD #GenerativeDesign #Engineering

AI for CAE: how can you rely on a model's predictions when making design decisions? We could talk about technical aspects of the question, looking at different ways to increase reliability of ML models, but I want to talk about something else. AI is just a tool. And it only becomes useful in hands of experts who know how to use it. It never was about blindly relying on AI to make decisions, it is about giving quick insights to the engineers and supporting them in the process. Think about CFD. What I have seen in the industry: - Different companies rely on different solver codes, turbulence models, convergence criteria etc - In any given simulation workflow, results heavily depend on the exact mesh structure (even with the same settings), solver code version and OS version (!) - Even after countless validations, CFD results can almost never be reproduced exactly in physical tests If there is anything you can be sure about in CFD, is that it's definitely wrong. "All models are wrong, but some are useful". Same goes for ML models in CAE. We need to educate engineers on how to extract value from this technology.

https://lnkd.in/e_8C3fnx It has been one year since we hosted the team from AI Marketplace for a discussion on the potential impact of GenAI and Large Language Models (LLMs) on product lifecycle management (PLM). At the time, we focused on applications in Requirements Engineering, Enterprise Search, Document Management, and Tech Docs. We discussed the importance of a robust digital thread strategy for realizing the potential of AI and the potential near- and long-term applications of AI and digital threads. AI continues to be a hot topic in our industry. CEOs expect engineering and PLM leaders to have an AI strategy. So, what has changed in a year? Join thought leaders from last year’s event as well as an expanded group of experts who will revisit AI’s impact on and potential for PLM and explore topics including: - The state of PLM copilots for assisting users with day-to-day activities - AI-assisted low-code development of new reports and applications - Augmenting the digital thread with AI-discovered relationships - Democratizing MBSE, advanced simulation, and system optimization - The future of generative design in engineering. Aras Corporation #MBSE #engineeringsimulation #ai #artificialintelligence #productdevelopment

The 2-minute video below illustrates how 'Document Comparison' works on Robotic Online Intelligence (ROI)'s Kubro(TM) platform. With the assistance from LLMs, you can quickly distil the difference between two PDF files (e.g. NVIDIA annual reports for 2023 and 2024), image-based PDFs (UK Companies House filings such as Deepmind Technologies' 2023 vs 2024 reports), or HTML documents (e.g. SEC filings such as 10-Q or 10-K). The challenge is how to efficiently (LLM token costs / time cost / relevance-accuracy) do it with larger documents, such as reports with 300+ pages. We find using our 'Regexy RAG' shortcut can work very well here, if the scope is two documents at the time, with some one-off setup that would serve similar cases, hence most suitable for the research process that involves ongoing, rather than ad-hoc, interrogation of similar-type documents. The 'Document Comparison' is one of several new tools in Kubro(TM) Document Explorer - a set of tools on the platform to assist analysts in dealing with larger unstructured documents, intended for data, research, and investment firms. #LLM #AI #Kubro #roboanalysts #intelligentautomation #research

If AI is the new “Wild West” then data becomes the new gold! Virtural Twin strategy fuels the next innovation wave. AI-Powered Virtual Twin Experiences: Transforming Innovation with Intelligent Insights for next Level Informed Decision Making! #GeneretiveAI #AIInnovation #VirtualTwinRevolution #AI-Products-R-Smarter #Desire2Inspire 🚀🤖