CoreWeave and Aston Martin F1 cut wind-tunnel-to-track prediction error by 50% with ML
capetorch · x · 2026-11-18
A CoreWeave blog post details its work with Aston Martin Aramco F1. In F1 aerodynamics, wind tunnel and track data never fully agree, and teams burn practice time diagnosing the gap.
Using precise downforce/drag measurements from the CoreWeave AI.R Tunnel at the AMR Technology Campus in Silverstone, the team trained an ML model that predicts the gap before the car turns a wheel, cutting the error in predicting on-track aero performance from wind tunnel data by 50%.
More from Companies & People
- Reported: Nvidia hires executive headhunter Heidrick & Struggles to find its next CEO — Polymarket · 2026-11-18
- Founder launches AI for Science startup training foundation models for multi-agent discovery — xennygrimmato_ · 2026-11-18
- Can Anthropic's IPO Succeed Without Nvidia's $10B Commitment? — IBDinvestors · 2026-11-18
- 'Mis-anthropic': AI circle memes Anthropic after Jensen Huang's ouster — beffjezos · 2026-11-18
- NVIDIA reportedly acquires Inception; diffusion model weights may go open source — mathemagic1an · 2026-11-18
- UCL Prof Konstantin Klemmer recruiting PhD students in geospatial foundation models, deadline Dec 3 — kklmmr · 2026-11-18