VJML / SYSTEMS LABLINUX / ARCHITECTURES / INFRASTRUCTURE / LOCAL AI

I build, explore, understand and reinvent systems.

I work with Linux, open-source software, alternative architectures, self-hosted infrastructure and local AI. This is where I document the projects and experiments born from that curiosity.

CRUX LINUX / ARM + RISC-V / SELF-HOSTING / LOCAL AI
Explore projects Enter the Lab

What I work on

Four areas connected by the same way of working: understanding systems, controlling infrastructure and learning by building.

01
CRUX Packaging System tooling Automation

Linux & Open Source

Linux has been the common thread behind most of my technical work. From maintaining and extending CRUX Linux to building packages, tools and infrastructure around it, I enjoy working close to the system.

02
ARM RISC-V SBCs Cross-compilation

Alternative Architectures

ARM and RISC-V are an important part of my experimentation, from adapting CRUX to different platforms to working with SBCs, native builds, cross-compilation and hardware with very different constraints.

03
Linux Containers Virtualisation Observability

Infrastructure & Self-hosting

My lab combines physical machines, virtualisation, containers, networking, storage, monitoring and automation to run services locally and remotely.

04
Local LLMs RAG Embeddings AI tooling

Local AI & Experimentation

I experiment with local language models, RAG, image generation, embeddings and AI-assisted workflows, with a particular interest in running them on infrastructure I can understand and control.

Featured projects

Some of the projects I currently work on or that form an important part of my lab.

01
Linux ARM RISC-V

CRUX

Work around CRUX Linux and its ports for alternative architectures, including package maintenance, tooling, automation and experimentation on ARM and RISC-V.

Explore CRUX projects
02
Python Streaming Automation

FluxTuner

A live streaming project built around continuous processing, automation and real-time workflows, with Brynse providing boundary-driven stream segmentation as an independent engine.

Explore FluxTuner
03
Local LLM RAG CUDA Self-hosted

Local AI Lab

A lab for experimenting with local models, RAG, embeddings, accelerated inference and AI tooling running on infrastructure under my own control.

Explore the lab
04
Image generation Local AI MLX

ai.vjml.es

An experimental image-generation service based on locally executed models, also used as a testbed for comparing models, resolutions and generation workflows.

Explore ai.vjml.es

Currently exploring

2026.10
RISC-V Native builds, toolchains and CRUX ports.
Local AI LLM inference, RAG, embeddings and local tooling.
Image models Local generation, MLX and model comparison.
Automation Infrastructure and AI-assisted workflows.
The Lab VJML / LAB

A place to build, experiment and understand systems.

VJML is not just a collection of finished projects. It is also where I document experiments, infrastructure, hardware, problems, solutions and things I am still learning.

Linux ARM RISC-V Self-hosting Local AI
→ Active