SSamu (samuphys)·Sep 16Implementing Kolmogorov-Arnold Networks as Dense GEMMs in JAX: A Neural QMC ExperimentReplacing B-splines with orthogonal polynomials and jnp.einsum in a Slater–Jastrow–backflow wavefunction for H2.10 min read·samuphys-lab.hashnode.dev
RRebecca Shuang Gao·Aug 23GeoWorld Now: Public Standards, Private Intelligence, One Scientific WorldSee how GeoWorld evolved into an open-core scientific AI platform with public standards, private intelligence, deterministic science, and provenance.5 min read·geoworld.hashnode.dev
SSarkis M·Jun 26Primer Vectors, J2, and Cones: Reading a Fuel-Optimal Maneuver Planner as Math, Not Just CodeHow koenig-damico-planner turns the Koenig & D'Amico (IEEE TAC 2020) algorithm into a finite-difference-verified Rust crate, and what a transcription typo in the published STM taught us about trusting the paper26 min read·railway-ts-pipelines.hashnode.dev
RRebecca Shuang Gao·Jun 23Why I’m Building GeoWorld: From Geoscience Language to a Learning Computable EarthGeoWorld connects natural-language geoscience questions with domain intelligence, scientific computing, uncertainty, and reproducible evidence.5 min read·geoworld.hashnode.dev
AAlberto Martinez morales·May 123 Bugs in My Exoplanet Simulation That Only Physics Could FindHow physical intuition saved my master's thesis simulation pipeline — and what that says about scientific thinking in tech.8 min read·physicsofdata.hashnode.dev
SSarkis M·May 9Izzo's Lambert solver, in your browserA WebAssembly Lambert solver for trajectory work that lives outside MATLAB and Python12 min read·railway-ts-pipelines.hashnode.dev
Zzubairakbar02·Apr 1Understanding the Hardware Behind an HPC ClusterLearn how an HPC cluster works by understanding its core hardware components including head node login node compute nodes GPU nodes memory optimized n4 min read·clustercraft.hashnode.dev
EEce Özen İldem·Mar 16CERN School of Computing Diaries - Day 1A reflection on Day 1 of the CERN School of Computing: profiling scientific software, neural scaling laws in AI, and parallel computing with RISC-V ve10 min read·codebeyondtheearth.hashnode.dev
PPietro Zanotta·Dec 29, 2024Learning Parametric Partial Differential Equations using Fourier Neural OperatorA variety of problems in applied science revolve about solving systems of parametrized partial differential equations. Such systems often exhibit complex and non-linear behaviour and mesh-based methods might therefore require an incredibly fine discr...8 min read·amm.zanotp.com
SSreya Ghosh·Jun 29, 2024Is it a fair game: Numerical vs AI?This blog unravels the depth of fundamental difference between numerical methods and Artificial Intelligence4 min read·phiwhyyy.hashnode.dev