Matteo Horvath, Building mesh.applied — efficient local LLM inference. Previously built mesh. in Budapest.

  • Blog
  • About
  • Projects
  • Search

I'm building mesh.applied: task-specialized open models and inference kernels so they run fast on local hardware. Before that I built mesh., a builder space in Budapest, from 0 to 35 people.

GitHub · LinkedIn · X

Why is communication so hard?

Jan 21, 2026

How I think about communication

mesh retro

Jun 6, 2025

Retrospective of our journey so far

Why you cannot get to PMF

Apr 18, 2025

things i am thinking about

Life in 10 years

Apr 18, 2025

matteo.getLatestPrediction()


© Matteo 2024 - 2026

GitHub · LinkedIn · X