Z80-μLM is a 2-bit quantized language model small enough to run on an 8-bit Z80 processor (via deejayy) — discussion

#ai #retrocomputing

Making end-to-end encrypted AI chat feel like logging in (via manuel) — discussion

#browsers #cryptography

ASK: Confidential computing for self-hosted communities (by koala)

#ask #security

Writing Windows 95 software in 2025 (via ralish) — discussion

#retrocomputing

hindsight: GitHub-style git activity visualizer for terminal (by chaosprint) — discussion

#rust #vibecoding

A Gentle Introduction to CUDA PTX (via LesleyLai) — discussion

#assembly #performance

Pidgin 3.0 Experimental 5 (2.94.0) has been released (by grim) — discussion

#c #release

Statically Linking PipeWire (via technetium) — discussion

#linux

corroded: Illegal rust (via ahobson) — discussion

#rust

A silly diffuse shading model (via LesleyLai) — discussion

#graphics

ASK: How do you secure access to your self-hosted/homelab services? (by tffr)

#ask #devops

2025: The year in LLMs (via mitsuhiko) — discussion

#vibecoding

Bokehlicious Selfies in Python (via azhenley) — discussion

#programming

I canceled my programming book deal (by azhenley) — discussion

#programming

HP-UX hits end-of-life today, and I’m sad (via classichasclass) — discussion

#historical #unix

Atmospheric Computing (via icy) — discussion

#distributed

Why Object of Arrays (SoA pattern) beat interleaved arrays: a JavaScript performance rabbit hole (via aapoalas) — discussion

#javascript #performance

A way to do an open and permissionless mesh network (by fiatjaf) — discussion

#distributed #networking

Reika: an opinionated dependency manager and build initiator tool (via amoroso) — discussion

#lisp

AI-generated content in WIkipedia - a tale of caution (via gnyeki) — discussion

#vibecoding #video