A text prompt that “almost” makes a bracket is the fastest way to lose an engineer’s trust. I sit down with Peggy Xia, co-founder and CEO of gNucleus, to find out what, besides almost making a bracket, gNucleus was up to.

Peggy's background is not that of a typical AI startup. She has had a career in the real mechanics of engineering AI, CAD automation and simulation workflows with Siemens. He wrote the code for Solid Edge's Synchronous Technology. At Google, she developed YouTube's most widely used recommendation model, which emphasizes video quality.

We unpack why text-to-CAD goes wrong in ways that feel unforgivable to practitioners: the model is forced to guess when it has not seen enough domain data, and that guessing manifests as hallucinations such as misplaced fillets or incorrect features. Peggy explains the training stack in practical terms: pretraining, post-training with reinforcement-style scoring, and fine-tuning on customer-specific CAD data, so a model can become genuinely strong in a narrow domain such as automotive motors or assemblies. We also talk about multimodal AI, what “sketch to CAD” could look like, and why converting meshes or photogrammetry outputs into clean, manufacturable parametric models is still one of the hardest problems in the pipeline.

From generative design to computational design, we challenge the idea that a cool-looking shape equals an engineering solution. Accuracy, tolerances, benchmarking, token cost, and runtime matter, especially when you want production-ready results, not just visuals.

If you care about the future of CAD, CAE, and manufacturing-grade AI, this conversation will sharpen your mental model of what’s possible now and what still needs breakthroughs.

Waiting months for “the next big CAD release” feels out of step with how fast AI is evolving, so we sit down with Darren Henry (SVP, PTC) and Cody Armstrong (Senior Director, Onshape AI Innovation) to explain what Onshape Labs is really designed to do. It is a broad, user-accessible early access program for what’s next in Onshape, where experimental features can ship faster than the normal three-week release cadence, get real feedback early, and then either graduate into the production cloud CAD product or get shut down quickly.

We also unpack the first Labs projects and why they matter to working engineers. The Omniverse Publisher, available via the Onshape App Store, focuses on robotics by moving CAD assemblies into NVIDIA Omniverse and Isaac Sim with far less friction. Because Onshape mates capture mechanical intent and degrees of freedom, teams can define joints and physical properties like stiffness and damping, then run kinematics and dynamics simulations in Isaac. The result is a tighter CAD-to-simulation loop for robotics design, physical AI workflows, and faster iteration between design changes and realistic motion behavior.

Then we go deep on FeatureScript MCP, which turns “custom features” into something far more accessible. Instead of relying on fragile text-to-CAD outputs, the approach is text-to-code-to-CAD: an LLM generates FeatureScript, inserts it, tests it, fixes errors, and iterates until it works. The best part for teams is that the LLM cost is mostly upfront, while the finished custom feature runs natively and can be shared across the organization without ongoing token usage. We cover demos, token realities, and why persistent memory could eventually capture company standards and tribal knowledge for the next generation of engineers.

“Make me a car” is an impressive demo until you ask where the braking hydraulics, controls, and safety logic went.

We let Viral Shah, CEO and founder of JuliaHub, and Chris Rackauckas, tell us about physical AI and its use for systems analysis.

Its a different AI story than the popular one: physical AI for engineers, where models must respect governing equations, compile, and validate against known test cases. Along the way we unpack why Julia, a programming language, was created and how it led to Dyad. Hint: to do systems analyses. Any system. Also how open source shaped its growth, and why that foundation matters when you want AI to do more than autocomplete code.

We then get concrete with Dyad, JuliaHub’s domain-specific language for systems modeling and multiphysics simulation. Rather than building another CAD tool, Dyad focuses on function over form, the system and subsystem level where real products live. That unlocks fast iteration in the engineering V-model: requirements, architecture, integration, and ultimately digital twin workflows, without forcing every engineer to become a full-time programmer.

The highlight is a demo where an agent ingests NASA HL-20 lifting body documents and aerodynamic data, generates a working systems model, runs a documented pitch-pulse test, and produces plots you can compare to the original validation figures. We also talk about the trust problem with AI and why physics-aware compilers, transparent artifacts, and test cases change the conversation from “wow” to “verify.” If you care about engineering simulation, systems engineering, agentic AI, and digital twins, subscribe, share this with a colleague, and leave a review with the tool you want AI to tackle next.

CAD should be the easiest tool in the room, yet it still feels like stepping into a cockpit packed with controls you have to relearn every time. We sit down with Mark, a longtime developer and former Siemens leader, to talk about why AI is racing ahead in some areas while product design and manufacturing still feel stuck and why the biggest blocker is not geometry, it is the interface.

We break down what large language models are genuinely great at today: procedural work, fast research, documentation, and especially coding. Mark explains why “vibe coding” and multi agent workflows are changing software development, pushing value toward architecture, domain experience, and good judgment. We also get real about the messy side: non deterministic outputs, security risks, and the stress of managing agents that never stop running.

Then we move to the physical world. Robots look impressive on stage, but on the factory floor speed, sensing, touch, and reliability matter more than demos. We talk about physical AI, why humanoid robots are both tempting and often impractical, and where near term wins actually live, like AI that helps operators troubleshoot CNC errors instantly or reduces time wasted searching documentation.

Finally, we connect the dots back to CAD, CAM, and CAE: automated drawings, design checking, natural language design, generative design exploration, and the emerging standards like MCP that aim to connect AI to engineering tools. If you care about the future of engineering software, AI for manufacturing, and the next generation of CAD interfaces, this conversation will sharpen your thinking.

Rare earth magnets, AI data center energy demand, and electrification are colliding in one place most people ignore: the electric motor. I sit down with Jarek Rzepecki from Monumo to get practical about what it takes to design motors and powertrains when costs, materials, and constraints can shift fast, and when “just optimize the motor” is never the whole story.

We dig into why system-level optimization matters, how a change to one component can cascade through the entire design, and why engineers can’t realistically brute-force the search space as parameters multiply. Derek explains how physics-informed AI, machine learning, and simulation can work together to explore designs faster, including approaches that reduce reliance on rare earth magnets while keeping performance targets intact. We also break down the motor landscape in plain terms: permanent magnet motors, wound rotor designs, and magnet-free reluctance motors, plus the real-world problem of torque ripple and what it does to noise, vibration, and durability.

Along the way, we connect the dots to robotics actuators, drones, generators, and the broader sustainability angle, because improving efficiency on the generation side and the consumption side can move the needle at global scale. If you like engineering software, multi-physics simulation, FEM, and the future of AI for engineering design, you’ll get a lot out of this interview.

Your wireless device can pass every isolated RF check and still disappoint in the real world. That’s the uncomfortable truth behind crowded stadium Wi-Fi, high-speed mobility, and the next wave of machine-type communication, and it’s exactly why I sat down with the team behind TSF Waves to unpack what “system-level wireless design” actually means.

We get into the hard split that’s held the industry back for years: RF and antenna engineers work inside electromagnetic theory and tools like ANSYS HFSS, while signal processing engineers live in algorithms, scheduling, decoding, and 3GPP-style resource allocation. For 5G and emerging 6G, especially at millimeter wave with large antenna arrays, those worlds collide. TSF Waves explains how they couple physics-based electromagnetic simulation inside the ANSYS Electronics Desktop ecosystem with a signal processing layer to produce system-level KPIs like channel capacity, block error rate, and usable spatial streams, so teams can evaluate hardware choices against real network performance.

We also talk about why edge AI, IoT, robotics, and V2X vehicle-to-everything connectivity are forcing order-of-magnitude jumps in data rates while power consumption stays a constraint. Then we explore their practical answer: a workflow-driven Python API and a “Wireless AI” agent that helps engineers run complex HFSS-based workflows without living in tedious code, while still understanding what they’re doing.

If you care about 5G, 6G, RF simulation, digital twins, and making wireless design faster and more reliable, subscribe, share this with an engineer on your team, and leave a review with the biggest wireless problem you want solved next.

We talk with John Harrington, co-founder of HighByte, about why factory-floor data stays invisible to the teams who need it most and how Industrial DataOps closes that gap. We explore contextualized data pipelines, the post-IoT architecture shift toward cloud data platforms, and why AI agents will force a new level of data quality and governance.
• Moving beyond “throw it over the wall” design and giving engineers real manufacturing feedback loops
• Defining Industrial DataOps and why context makes raw OT data usable
• Handling messy realities across MES, ERP, historians, inspection systems, files, and streaming telemetry
• Avoiding data swamps by standardizing, governing, and observing data pipelines at scale
• Using no-code tooling to build and maintain pipelines without relying on programmers
• Filtering and sampling data based on use case, frequency needs, and event triggers
• Preparing for AI agents as massive new consumers of shop floor data
• Realistic talk on AI and jobs, focusing on better work through better signal detection

We talk with Juan Carlos Santamaria about how AI in engineering has evolved from rule-based robotics to modern systems that perceive job sites and help machines make better decisions. We dig into Trimble’s push from AI perception to operator assist and what it will take for engineers and operators to trust AI in the field and in design tools.
• Juan Carlos’s PhD-era view of AI as a multidisciplinary field
• Planning versus reactive robotics and why brittle plans fail
• AI for perception on construction sites using point cloud images and video
• Turning recognition into jobsite semantics like cycles and bucket loads
• The shift toward decision making with operator assist in the cab
• Autonomy in mining versus the realities of safety policy and adoption
• Why trust builds faster when operators can experience the system
• Zero tolerance expectations for machines compared with human error
• Point cloud segmentation today and what engineers want next
• Natural language interfaces that execute software commands from prompts
• SketchUp and 3D Warehouse visual search and AI-assisted edits
• Interoperability across tools and the “Tower of Babel” problem
• Planning for unknown unknowns when digging into existing infrastructure
• How AI work gets prioritized across product teams
• Why kids and professionals should learn AI as a tool for thinking clearly