Skip to the card

Card 656 of 9752026-09-18 issue

Observed arrival · 2026-09-18

PINNeAPPle Labs Wants Physics AI to Show Its Work

pinneapple.org Visit website
Editorial interest 86/100 Selection signal · not a rating of the site

A physics-AI lab presenting verification tools, neural simulation surrogates, a live river digital twin, and open-source PINN software.

Landing page captured for the 2026-09-18 issue.
For
Engineers evaluating physics-informed AI
Worth noticing
The homepage reports 4.46% average and 7.35% worst-case agreement with independent 3D FEA across seven heatsink designs.

Field notes

The project separates its public software foundation from commercial demonstration products. Listed components include PINNeAPPle for PINN, FNO, DeepONet, GNN, solver, and digital-twin work, plus pinnfactory for symbolic PDE definitions using SymPy and automatic differentiation in PyTorch. Its verification example emphasizes per-check reporting: residual, convergence, calibration, and an explicit “not run” state when a check does not execute. A confidential industrial model is cited as an internal guardrail test.

Observed signals

Read the marks

Editorial observations of this landing page, not a rating.

OpenPublic substance visible
PrettyNotable craft visible
ProPolished or operationally mature
NicheUnusually specific use

One card from the complete issue

Please Unpack 378 Paintings

344,538 arrived 1,000 judged 975 catalogued Enter the complete issue
pinneapple.org

Landing page observed 2026-09-18. The live site may have changed.