Skip to content
View CastelDazur's full-sized avatar
39 followers · 47 following

Achievements

Achievement: Starstruck

Achievements

Achievement: Starstruck

Block or report CastelDazur

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
CastelDazur/README.md
CastelOS Boot Sequence

I build AI systems that execute, not just chat. Local-first. Governed. Hardware-aware. Real.

Python TypeScript CUDA Docker Linux

What I do

I design and build CastelOS — a local-first AI execution system that turns tasks into governed runs with real artifacts and evidence. Not another wrapper around an API. A full system: from GPU routing to policy enforcement to domain-specific knowledge packs.

Everything runs on one workstation I assembled myself. No cloud dependencies. No scattered SaaS. Just execution.

Things you can run

The reusable pieces live in their own repos. None of them need CastelOS to be useful.

  • awesome-local-ai is a curated list of tools for running models on your own hardware.
  • gpu-memory-guard stops one process from taking the whole GPU. pip install gpu-memory-guard
  • llm-judge-jury puts several models to a vote on an output instead of trusting one judge.
  • qlora-single-gpu-playbook is the set of guards that kept my QLoRA runs alive on a single card.
  • merge-quantize-keep-mtp merges two LoRA adapters by rank, then verifies the MTP heads survive GGUF quantization.

CastelOS Runtime Lifecycle


castel.studio   LinkedIn   X

Pinned Loading

  1. castelos-public castelos-public Public

    CastelOS public artifacts — principles, architecture insights, and build-in-public content

    Mermaid 2

  2. awesome-local-ai awesome-local-ai Public

    A curated list of tools, frameworks, and resources for running AI locally — inference, RAG, orchestration, monitoring, and more.

    20 8

  3. gpu-memory-guard gpu-memory-guard Public

    CLI tool to check GPU VRAM before loading AI models. Prevent OOM crashes.

    Python 10

  4. llm-judge-jury llm-judge-jury Public

    A jury of independent LLM judges for dataset QA. Score instruction/response pairs, keep the consensus.

    Python

  5. merge-quantize-keep-mtp merge-quantize-keep-mtp Public

    Merge two LoRA adapters by rank-concatenation and verify speculative-decoding (MTP) heads survive GGUF quantization.

    Python

  6. qlora-single-gpu-playbook qlora-single-gpu-playbook Public

    Guards and helpers for QLoRA on a single GPU: VRAM gating, resume without the hang, checkpoint repair, preflight checks.

    Python