Tag: #jax
Writing on GPUs, LLMs, MLOps, Kubernetes — and mindset · 5 posts
Modern Statistical Computing 2026 Complete Guide - R 4.5 · Posit RStudio · Stan · Pyro · NumPyro · Brms · JAX · Tidyverse · data.table · Polars · Marimo Deep Dive
In 2026 statistical computing has two large currents running in parallel. On one side R 4.5 (April 2025) and Posit (formerly RStudio) commercial stack are cementing the reproducible-research standard. On the other side J
2026-05-16 · 23 min read #statistical-computing#r-language#posit#rstudio#stanModern Fortran & Scientific Computing 2026 Deep Dive - Fortran 2023, LFortran, gfortran, NumPy, SciPy, JAX, Julia, APL, J, K
A complete topographic map of scientific computing in 2026. Fortran, born in 1957, is not dead — through Fortran 2023, LFortran 0.50, gfortran 14, Flang/LLVM, NVIDIA HPC, ifx, fpm, stdlib, OpenMP 5.2, MPI 4.1, CoArray, a
2026-05-16 · 31 min read #fortran#scientific-computing#lfortran#gfortran#numpyMath and Scientific Computing Tools in 2026 — Mathematica / MATLAB / Maple / SageMath / Julia 1.11 / R + Posit / JAX / GeoGebra / Desmos / GAP / Macaulay2 Deep Dive
The math and scientific computing landscape in 2026 is no longer dominated by one or two tools. Wolfram Mathematica still owns premium symbolic computation. MATLAB and Simulink remain the engineering industry standard. M
2026-05-16 · 29 min read #math#scientific-computing#wolfram-mathematica#matlab#simulinkDistributed Training & GPU Infrastructure 2026 Deep-Dive — DeepSpeed, FSDP2, Megatron-LM, Ray Train, JAX, TorchTitan, Blackwell GB200, MI325X, TPU v5p
A comparison of DeepSpeed, FSDP2, Megatron-LM, Ray Train, JAX, TorchTitan, and Composer — plus NVIDIA Blackwell GB200 NVL72, AMD MI325X, Intel Gaudi 3, AWS Trainium 2, and Google TPU v5p/v6e Trillium. 3D parallelism, ZeR
2026-05-16 · 14 min read #distributed-training#deepspeed#fsdp#megatron-lm#rayGoogle TPU Deep Dive: How Systolic Arrays Solve Matrix Multiplication Perfectly
A complete technical breakdown of how Google's Systolic Array achieves extreme efficiency for matrix multiplication. From INT8 inference and bfloat16, to XLA compiler optimizations and TPU Pod distributed inference - wit
2026-03-18 · 15 min read #tpu#google#systolic-array#model-serving#jax