How EGI generates interactive assembly instructions from CAD

Extracting critical data points from CAD drawings like part names, material specifications, dimensions, and revision details no longer has to be a manual, time-consuming task. With AI, it’s now fast, accurate, and effortless. AI systems can automatically read and capture information from complex CAD files, standardize formats, and organize data for easy access. This means fewer manual errors, quicker design validation, and improved collaboration across engineering teams. From faster documentation to smarter design insights, AI is reshaping how CAD data is managed and utilized. Read the full blog to explore how automated extraction enhances precision and productivity: https://lnkd.in/gXS3j3Zg #AI #CAD #Engineering #Automation #DataExtraction #DesignAutomation #SmartManufacturing #ProductDesign #DigitalEngineering #AIinEngineering #Innovation #TechEfficiency #EngineeringProductivity #AutomatedWorkflows #AIDrivenDesign

𝗠𝗢𝗗𝗘𝗥𝗡 𝗦𝗢𝗙𝗧𝗪𝗔𝗥𝗘 𝗗𝗘𝗩𝗘𝗟𝗢𝗣𝗠𝗘𝗡𝗧 𝗟𝗢𝗢𝗞𝗦 𝗟𝗜𝗞𝗘 𝗛𝗔𝗥𝗗𝗪𝗔𝗥𝗘 𝗗𝗘𝗩𝗘𝗟𝗢𝗣𝗠𝗘𝗡𝗧 AI is changing that fast — and for good. My journey in tech actually started with hardware. One of my earliest experiments was mounting motors and batteries on a piece of plywood. It might sound silly now, but that was probably the most joy I’ve ever felt while building something. The moment the robot moved forward for the first time was pure magic. But my journey into software actually began from a small frustration with hardware. Every time I finished designing a circuit board and sent it for printing, it would take two to three weeks to come back. Then during testing, I’d realize I’d made one tiny mistake and I’d have to fix and send it again, waiting another three weeks. You could call it a skill issue, but it wasn’t just me. There’s an entire industry built around preventing those mistakes — the world of CAD design and simulations. You build your whole product on a computer, run simulations, check the fit and finish, and only when you’re absolutely sure, you move to manufacturing. The reason is simple: printing a circuit board or machining a part takes time, costs money, and mistakes can set you back by months. That was the biggest pull for me toward software. Because I could make a hundred mistakes and instead of losing weeks, I’d lose a few minutes — and then be back on track. I could keep changing and improving my idea without feeling bad for not predicting everything upfront. Over time though, as I started building more serious software — things used by lakhs of people now — I realized that software development has started to look a lot like hardware again. Instead of 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝗱𝗿𝗮𝘄𝗶𝗻𝗴𝘀, we have 𝘄𝗶𝗿𝗲𝗳𝗿𝗮𝗺𝗲𝘀. Instead of 𝗖𝗔𝗗 𝗱𝗲𝘀𝗶𝗴𝗻𝘀, we have 𝗙𝗶𝗴𝗺𝗮 𝗱𝗲𝘀𝗶𝗴𝗻𝘀. Instead of 𝗽𝗵𝘆𝘀𝗶𝗰𝘀 𝘀𝗶𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻𝘀, we have 𝗶𝗻𝘁𝗲𝗿𝗮𝗰𝘁𝗶𝘃𝗲 𝗽𝗿𝗼𝘁𝗼𝘁𝘆𝗽𝗲𝘀. And all of this happens 𝑏𝑒𝑓𝑜𝑟𝑒 the first line of code is written. Why? Because engineering is expensive. Developers have limited time and focus. Writing good code still takes effort and craftsmanship. But AI is changing that. I want to call this new phase “𝗧𝗵𝗲 𝗖𝗹𝗮𝘆 𝗠𝗼𝗱𝗲𝗹𝗹𝗶𝗻𝗴 𝗘𝗿𝗮 𝗼𝗳 𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁.” A sculptor starts with a lump of clay and a vision in mind. He builds the base first, then slowly refines and adds 𝑓𝑒𝑎𝑡𝑢𝑟𝑒𝑠 as the form takes shape. It’s hands-on, iterative, and you can see the vision take shape as you go. That’s where I think software is heading — an era where creators can shape ideas directly. With AI, 𝘁𝗵𝗲 𝗰𝗼𝘀𝘁 𝗼𝗳 𝗰𝗿𝗲𝗮𝘁𝗶𝗼𝗻 𝗶𝘁𝘀𝗲𝗹𝗳 𝗶𝘀 𝗰𝗼𝗹𝗹𝗮𝗽𝘀𝗶𝗻𝗴, and so is the gap between idea and execution. 𝗠𝗮𝘆𝗯𝗲 𝗼𝗻𝗰𝗲 𝗮𝗴𝗮𝗶𝗻, 𝘀𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗰𝗮𝗻 𝘁𝗿𝘂𝗹𝘆 𝗳𝗲𝗲𝗹 𝗹𝗶𝗸𝗲 𝘀𝗰𝘂𝗹𝗽𝘁𝗶𝗻𝗴. Vishranth Suresh AssetPlus #AI #ProductDevelopment #SoftwareDevelopment #FutureOfWork

At AssetPlus, we’ve always believed every great product begins with experimentation and vision. As our Co-founder Awanish Raj shares, it feels like we're entering The Clay Modelling Era of Software Development - where ideas can be shaped, tested, and refined in real time. What once defined hardware prototyping is now transforming software creation, powered by AI. The distance between concept and execution is collapsing. It’s a new age of building - fast, fluid, and full of possibility.

How AI Is Quietly Fixing the GD&T Problem in Mechanical Design Most design reviews still depend on manual checks — zooming into drawings, verifying datums, and cross-referencing standards line by line. It works, but it’s painfully slow. And even the best engineers miss details when deadlines close in. If you’ve been in design long enough, you’ve seen it all: * Missing or duplicated datums * Conflicting tolerances across views * Drawings that pass peer review but fail inspection These mistakes cost more downstream than anyone likes to admit. Now, AI is starting to close that gap — not by replacing engineers, but by enforcing the precision we always intended. An AI agent can now: 🔹 Parse GD&T symbols directly from 2D drawings or 3D annotations. 🔹 Cross-check each feature against ASME Y14.5 or ISO 1101 standards. 🔹 Flag incomplete callouts, conflicting tolerances, or missing datums instantly. 🔹 Compare new drawings with approved ones to maintain consistent tolerance logic. Instead of spending half an hour validating a single drawing, engineers can focus on design intent, while AI quietly ensures compliance and consistency behind the scenes. AI isn’t replacing judgment — it’s scaling it. That’s the real transformation happening in mechanical design today. Great to see leaders like PTC, Dassault Systèmes, Siemens Digital Industries Software, and Hexagon Manufacturing Intelligence AB leading the charge toward AI-driven design intelligence. How soon do you think this will become standard practice across the industry? Comment your thoughts! DM me if you want more information about building the AI-driven GD&T rule checking agent (#n8n) #MechanicalEngineering #DesignAutomation #AIforEngineers #GDandT #DigitalManufacturing #PLM #EngineeringDesign #DesignReview #SmartManufacturing #CATIA #NX #Creo #SolidWorks

Is AI pushing us toward a higher level of mediocrity? I’ve been in automation for over 35 years. Back then, we didn’t have access to the incredible sensors, servos, and off-the-shelf solutions available today. Just finding a sensor that could do the job required deep research, creativity, good vendors on speed dial and your trusty ole Thomas Register. Most young engineers today wouldn’t know where to begin without the internet and that’s not a knock, it’s just how much the landscape has changed. Which brings me back to the question: is AI nudging us toward mediocrity? In the world of custom automation, it’s the engineers, programmers, and technicians who make the difference. The human element is what sets the true top tier automation companies apart. Back in the drafting table days I could tell who designed something without even looking at the title block, you knew from line weights, dimensioning style and attention to detail. A-CAD took most of this away then SolidWorks finished it off. When you become an expert at your craft you could pick up a 2D drawing and after a little time studying it you could visualize it in 3D. If AI becomes the primary design tool, what will differentiate one company from another? Sure, there will still be human input, but I can’t help but feel that things might start to look and feel the same. Maybe I'm wrong but I can't help but to feel like it will become harder for companies to differentiate themselves. Would love to hear your thoughts.

𝗕𝗲𝘆𝗼𝗻𝗱 𝗖𝗼𝗺𝗽𝗹𝗶𝗮𝗻𝗰𝗲 — 𝗧𝗵𝗲 𝗙𝘂𝘁𝘂𝗿𝗲 𝗼𝗳 𝗚𝗗&𝗧 (𝗠𝗕𝗗) 𝗧𝗵𝗲 𝗳𝘂𝘁𝘂𝗿𝗲 𝗼𝗳 𝗚𝗗&𝗧 𝗶𝘀 𝘀𝗺𝗮𝗿𝘁, 𝗶𝗻𝘁𝗲𝗿𝗮𝗰𝘁𝗶𝘃𝗲, 𝗮𝗻𝗱 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗲𝗱. When I first learned GD&T, it was all about checking boxes. Symbols, datums, tolerances — all tightly bound to compliance. Back then, the goal was simple: “Make sure the part passes inspection.” But over the years, as I’ve worked across global teams — from design engineers in Germany to machinists in India to quality leads in the U.S. — I’ve realized something deeper. GD&T isn’t just a language of compliance. It’s a language of collaboration. Every symbol on a drawing carries an intent — not just for tolerance, but for communication. When done right, it bridges the gap between what the designer imagines, what the machinist produces, and what the inspector validates. That alignment is priceless in a global supply chain. And now, we’re entering a new era — one where GD&T is no longer just 2D ink on a print. It’s becoming digital, data-driven, and model-based. 𝗜𝗻 𝗮 𝗠𝗼𝗱𝗲𝗹-𝗕𝗮𝘀𝗲𝗱 𝗗𝗲𝗳𝗶𝗻𝗶𝘁𝗶𝗼𝗻 (𝗠𝗕𝗗) 𝗲𝗻𝘃𝗶𝗿𝗼𝗻𝗺𝗲𝗻𝘁, 𝗚𝗗&𝗧 𝗹𝗶𝘃𝗲𝘀 𝗱𝗶𝗿𝗲𝗰𝘁𝗹𝘆 𝗶𝗻 𝘁𝗵𝗲 3𝗗 𝗺𝗼𝗱𝗲𝗹. 𝗡𝗼 𝗺𝗼𝗿𝗲 𝗮𝗺𝗯𝗶𝗴𝘂𝗶𝘁𝘆 𝗯𝗲𝘁𝘄𝗲𝗲𝗻 𝗖𝗔𝗗 𝗮𝗻𝗱 𝗖𝗠𝗠. 𝗡𝗼 𝗺𝗼𝗿𝗲 𝗺𝗶𝘀𝘀𝗲𝗱 𝗰𝗮𝗹𝗹𝗼𝘂𝘁𝘀 𝗼𝗿 𝗺𝗶𝘀𝗶𝗻𝘁𝗲𝗿𝗽𝗿𝗲𝘁𝗲𝗱 𝗱𝗮𝘁𝘂𝗺𝘀. The future of GD&T is smart, interactive, and integrated. AI-assisted tools are already helping designers auto-suggest tolerances based on functional intent. CMM software can read the same semantic GD&T data that design tools produce — reducing interpretation errors to near zero. But technology alone isn’t enough. The human element still matters most. Because at its core, GD&T is a conversation between people — expressed through symbols. If teams don’t share the same understanding, even the smartest software won’t fix it. The next leap in GD&T isn’t about tighter tolerances. It’s about tighter understanding. Imagine a world where design, manufacturing, and inspection teams — across continents — speak the same geometric truth fluently. Where tolerance isn’t just a number, but a shared intent. Where precision isn’t inspected in, but designed in from the start. That’s where we’re headed. In global engineering, the companies that master this “language of precision” won’t just make better parts —they’ll build stronger partnerships. What’s your take on the digital future of GD&T? Are you already exploring model-based GD&T or AI-assisted tolerance analysis in your workflows? #GDandT #ModelBasedDefinition #DigitalEngineering #PrecisionManufacturing #AIinEngineering #Quality4_0 #GlobalEngineering #AutomotiveEngineering #Aerospace #EngineeringLeadership "Image Credit To Solidworks"

The future of design is intelligent. AI is transforming how we use CAD tools from automating repetitive modeling tasks to generating complex geometries in seconds. Imagine designs that evolve based on real-world constraints, material data, and performance insights — all powered by AI-driven algorithms. 💡 With AI integrated into CAD, engineers can now focus more on innovation and creativity rather than manual drafting. The next era of design isn’t just about Computer-Aided Design — it’s about Cognitive-Aided Design. #AI #CAD #DesignEngineering #Innovation #SolidWorks #Automation #FutureOfDesign

💡 When AI Learns to Design I’ve been experimenting with something that could change how we think about CAD automation. We took a simple video of a user working in CAD, selecting elements, adding supports, measuring structures, and fed it into Qwen2.5-VL-7B-Instruct, a multimodal AI model that understands both visuals and context. Instead of captions, it produced structured, machine-readable instructions describing what the user was doing, line by line, tool by tool. Here’s a snippet: { "command": "add_support_points", "parameters": { "points": 4 }, "tool_location": "ribbon_tab/modify", "step_description": "Add support points to the scaffolding.", "automation_instruction": "add_support_points(4)" } That means the AI isn’t just seeing, it’s learning the workflow. We’re now able to convert a CAD demonstration video into an instruction set that could be executed automatically. Think of it as “Demonstrate to Automate”, where the system learns by watching a human designer. No macros. No code. Just human expertise, translated into logic. If this evolves as we expect, it could enable: Reusable automation templates built from design videos Platform-agnostic CAD learning (AutoCAD, Tekla, Revit, etc.) AI co-designers that replicate company standards with zero scripting This is early work, but it’s real. We may still be scratching the surface, but I genuinely think we’re onto something big here. Would love to hear your thoughts - How far do you think this approach could go in transforming design workflows? #AI #CADAutomation #VersalenceAI #AgenticAI #AIDesign #Qwen #DesignTechnology #Innovation #Kaggle

What if your words could build your ideas? Describe a part, a product, or a design — and within seconds, it becomes a 3D CAD model. This isn’t the future. It’s already transforming how engineers and designers work. Here are 5 AI-powered tools leading this transformation 👇 🧠 Zoo Design Studio (Text-to-CAD) – Translates natural language into precise 3D CAD models. Perfect for rapid prototyping and creative design iterations. 🔗 zoo.dev/text-to-cad 🧩 CADScribe – A free and intuitive AI that turns written ideas into CAD sketches. Great for students, hobbyists, and early product concepts. 🔗 cadscribelabs.com ⚙️ Leo AI (Engineering Copilot) – Your intelligent teammate for design, simulation, and documentation. Integrates seamlessly into modern engineering workflows. 🔗 getleo.ai 🧱 Vondy AI CAD Generator – Converts simple text inputs into 2D or 3D models, saving hours of repetitive design work for mechanical engineers. 🔗 https://lnkd.in/gaha4eqq 📐 insMind AI CAD Drawing Generator – Instantly generates accurate CAD drawings and diagrams from plain text or quick concept notes. 🔗 https://lnkd.in/gmSbPSUX 💡 AI isn’t here to replace designers — it’s here to enhance imagination, precision, and speed. Those who learn to collaborate with AI will define the next generation of innovation. #AICAD #TextToCAD #MechanicalDesign #EngineeringInnovation #GenerativeDesign #AIinEngineering #FutureOfDesign #ProductDevelopment #DesignAutomation

The AI revolution in design isn’t coming — it’s here. This article explores how SOLIDWORKS and the 3DEXPERIENCE platform are incorporating AI features today: generative geometry, constraint suggestion, error flagging, and variant exploration. These tools are augmenting designers, reducing toil, and accelerating ideation cycles. If your team uses CAD and you’re wondering how AI fits into your roadmap, this is a useful snapshot of where things stand. Read more: https://lnkd.in/eTg6WxXE