Show HN: SIMD Agent – AI that runs OpenFOAM simulations from natural language
SIMD Agent is an AI-driven physics simulation tool that allows users to run OpenFOAM simulations using natural language descriptions. It simplifies complex computational fluid dynamics tasks by automatically selecting solvers and diagnosing failures. The open-source platform supports various turbulence models and provides live streaming of simulation results.
- ▪The SIMD Agent can interpret natural language inputs to configure physics simulations.
- ▪It features a self-healing mechanism that retries simulations after diagnosing solver failures.
- ▪The tool supports multiple turbulence regimes and is compatible with various LLM providers.
Opening excerpt (first ~120 words) tap to expand
SIMD Agent AI-native physics simulation agent. 🎯 What it does Describe the physics you want to understand — flow, heat transfer, buoyancy, pressure drop, mixing — together with a mesh of the physical domain, and the agent picks the governing equations and turbulence model, configures discretization schemes, properties, and boundary conditions, submits the case to an OpenFOAM solver, streams residuals and flow fields back, and self-heals on solver failure. The result: design-decision support — questions that usually take a CFD specialist a week ("does this U-bend overheat at 500 K?", "what's the pressure drop through the regasifier?") become a paragraph of intent plus a geometry.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.