nTop is the engineering design platform built to increase design throughput and systematically explore complex design spaces.
New York, USA
Joined May 2015
- Tweets715
- Following466
- Followers6.6K
- Likes534
Pinned Tweet
nTop is featured in @nvidia's press release on agentic engineering.
With NVIDIA and @JetZeroAero, we're building an autonomous workflow around a blended wing body.
Read the fully story here: hubs.li/Q04jzhMW0
nTop retweeted
CAD+AI NYC
@brad_rothenberg hosted us at @nTopology last night, @earthtojake presenting text-to-cad, @__spigach__ talks structural engineering + AI and Will from Variant3D brought a knitted blender monkey head!
Iteration speed, weight optimization, and structural reliability are usually what separate a team that places from a team that wins. And the tools most teams start with tend to slow that process down right when it matters most, during early trade studies and detailed component design.
Student competition teams can access free educational licenses to power their projects throughout the season.
Request an EDU license here: ntop.com/education/
Our summer interns designed, built, and flew a blended wing body aircraft during their internship.
"Our nTop master files were fully parameterized. Any damage sustained during field testing was an opportunity rather than a setback. We could tweak part thickness or pin alignment, export a fresh mesh, slice, and re-print replacement sections in a matter of hours."
New display styles and a refreshed color palette in nTop.
UI. Colors, buttons, menus, modals, and viewport styling have all been refreshed.
We've also added two new display styles:
Shiny Plastic and Brushed Metal.
Also the color palette has been updated with richer, higher-contrast colors and new default color assignments, so your designs are easier to read and present.
By combining nTop’s implicit modeling engine, automated execution via nTopCL, and aerodynamic solvers including AVL and AeroSandbox.
One of our intern progressed from high-dimensional design space exploration to a physical, flight-tested aircraft that successfully demonstrated the gentle handling characteristics targeted by the design.
"Closing the Loop Between Implicit Geometry and Flight Physics"
resources.ntop.com/resources…
nTop retweeted
astra’s building the jet while I’m flying on a jet
I told it to get to PDR on a 20lbf class small jet in ntop and the only manufacturing processes available are metal forming and casting, use standard fasteners: it built out a set of requirements, generated the assembly in ntop, output a sysml 2.0 model of the system and ran initial analysis, and created unit tests to check against the requirements, it’s not at PDR yet, but still unbelievable an LLM can do this
Also its ntop model building skills are impressive.
JetZero’s nTop model of the Z4 is fully parametric, encoding the entire aircraft as a single interconnected workflow: outer mould line, structural wing box, air foil sections, nacelles, winglets, landing gear and interior volumes.
When parameters change, the model regenerates correctly and the output is simulation-ready every time.
nTop retweeted
more from astra, here's an inline 6 sizing study it ran after building the parametric model in ntop and figuring out what design variables to modify, again obv still not ready to manufacture of course, but this is still a step change in an LLM understanding 3D
Our summer interns designed, built and flew airplanes with different configurations.
Below is our first flight test of the ducted fan, blended wing body aircraft called "Vlad the Inhaler" (you can download the model here - community.ntop.com/#/detail/…).
The front landing gear didn't survive the first takeoff attempts, but that didn't stop us!
nTop retweeted
Guénaël on our european team built a CCA design explorer that runs openfoam straight from ntop
ntop ML orchestrates the sweep while our openfoam plug-in is used for implicit interop, all runable / tunable via LLMs and packaged up into a design explorer dashboard for us humans to review
Intact.Simulation for nTop connector runs directly on implicit geometry, fully compatible with nTop blocks.
Supports structural, thermal, and modal analysis, from quick design checks through higher-fidelity evaluation as a design matures.
No conversion, no simplification, no mesh.
Training a physics AI model takes thousands of CFD-ready geometry variants, generated fast enough to make it worthwhile.
nTop and Luminary generated close to 5000 variants together in a few weeks. With a physics model trained into the loop, engineers change the geometry and watch lift and drag update immediately.