coderredlab

https://github.com/coderredlab/runNburn

runNburn is an Apache-2.0 Rust inference engine for quantized GGUF models that are too big for your fast memory.The core idea: weights stay file-backed (mmap), host residency stays under an explicit byte budget (--ram-budget), and GPU caches are sized from detected free/total VRAM — never from device-name presets. There is no conversion step, no sidecar cache files, no silent requantization....

https://github.com/coderredlab/proroot

I built a drop-in replacement for proot that runs full Linux userspace (Node.js, Python, Git, Chromium) on Android without root.proot uses ptrace — 2 context switches per syscall. proroot uses LD_PRELOAD + binary patching — 0 context switches, in-process path translation.It powers andClaw, an app that turns Android phones into AI agent gateway servers running OpenClaw with a real Chromium...

https://play.google.com/store/apps/details?id=com.coderred.redterm&hl=en_US

Hi HN,I built this because using Claude Code or Codex CLI on a remote server from a phone was awkward whenever I needed to send an image.Typing commands on mobile is manageable, but moving an image from the phone into the remote workflow usually means saving files manually, uploading them separately, then finding the path again. That breaks the flow every time.So I made redTerm, an Android SSH...

https://github.com/coderredlab/andClaw

Hi HN,I built an Android app that bundles a minimal Linux environment and runtime to run OpenClaw locally on-device.The motivation was simple: I didn’t want to spin up and maintain cloud VMs just to host an AI gateway. Modern Android phones are capable enough, so I experimented with packaging the runtime directly inside the app using proot.The app installs a bundled Linux rootfs, runs the...