Skip to content
Infrastructure

Homelab

24 self-hosted services, 80+ automated workflows, and a local AI stack. All running on-premise. No cloud dependencies, no third-party data processing.

Architecture

Homelab network topology showing TerraMaster NAS with 40+ Docker containers, Mac Studio for local AI and compute, Home Assistant Green, and capture devices connected over local network
Capture
Omi WearablereMarkableRSS FeedsOura RingDocuments
Processing
n8n (80+ workflows)LM Studio (local LLM)WhisperX (speech-to-text)Paperless (OCR)
Knowledge
BookStack (wiki)Obsidian (notes)Qdrant (vectors)AnythingLLM (RAG)

Hardware

NAS / Docker Host

TerraMaster F2-425 Plus

Intel N150, 16 GB RAM, ZFS

40+ Docker containers

AI / Compute / Dev

Mac Studio

M3 Ultra, 96 GB RAM, OrbStack

LM Studio, Qdrant, AnythingLLM, WhisperX

Home Automation

Home Assistant Green

Dedicated HA appliance

Home Assistant + voice satellites

Network

GL.iNet Flint 3

Wi-Fi 7 router

AdGuard Home, WireGuard

Services (24)

Automation2AI4Knowledge5Media2Security4Monitoring2Tools5
n8nTrueNAS

Workflow automation: 80+ workflows handling integrations, pipelines, and scheduled tasks

Home AssistantHA Green

Smart home control with voice satellites and device automation

LM StudioMac Studio

Local LLM inference (Mistral Small 3.2 via MLX) with OpenAI-compatible API

WhisperXMac Studio

GPU-accelerated speech-to-text with speaker diarization for transcription pipelines

AnythingLLMMac Studio

RAG interface for querying documents with conversational search (768-dim embeddings)

QdrantMac Studio

Vector database powering semantic search across all stored content (768 dimensions)

Paperless-ngxTrueNAS

Document ingestion with OCR, tagging, and full-text search

BookStackTrueNAS

Internal wiki for journals, runbooks, and reference documentation

ObsidianMac Studio

Primary note-taking and knowledge management via Markdown

FreshRSSTrueNAS

RSS aggregation with AI-scored relevance filtering

LinkwardenTrueNAS

Bookmark manager with full-text archival and tagging

PlexTrueNAS

Media server for movies, TV, and music libraries

ImmichTrueNAS

Self-hosted photo management with facial recognition and search

CipherGateTrueNAS

Machine-to-machine secrets store: encrypted at rest, scoped per consumer, audit-logged on every read. Backs the MCP servers and automation jobs

VaultwardenTrueNAS

Bitwarden-compatible password manager for human logins, fully self-hosted

WireGuardRouter

VPN for secure remote access to all internal services

CloudflaredTrueNAS

Cloudflare tunnels for externally exposed services

Uptime KumaTrueNAS

Service health monitoring with alerting via Pushover

BeszelMac Studio

Real-time host and container metrics with alerting

Stirling-PDFMac Studio

Local PDF processing: merge, split, convert, OCR

Resilio SyncTrueNAS

Peer-to-peer file sync across reMarkable and workstations

CouchDBTrueNAS

Sync backend for Obsidian LiveSync: the vault replicates across devices through it

RustDeskTrueNAS

Self-hosted remote desktop access

Actual BudgetTrueNAS

Privacy-first budgeting and financial tracking

Automation Highlights

80+ n8n workflows handle recurring tasks, data pipelines, and infrastructure maintenance.

Morning Briefing

Daily summary delivered by email: weather, calendar, tasks, and Oura ring sleep data, generated by the local LLM.

Transcription Pipeline

Omi wearable captures conversations, WhisperX transcribes locally with speaker diarization, n8n classifies content and routes action items to Todoist.

Personal Knowledge Management Pipeline

Notes from Obsidian, scans from reMarkable, and documents from Paperless are processed, embedded, and indexed for semantic search.

Self-Healing Infrastructure

Downtime triggers AI triage, automatic restart attempts, and rollback if health checks fail. Alerts sent via Pushover.

Docker Update Pipeline

Weekly automated container updates with pre-update snapshots, health checks, and automatic rollback on failure.

RSS AI Digest

FreshRSS feeds scored by relevance, top articles summarized by the LLM, and a daily digest delivered by email.

AI Stack

Local LLM Inference

LM Studio serves Mistral Small 3.2 on the Mac Studio's M3 Ultra via an OpenAI-compatible API. All AI workflows (classification, summarization, briefings) route through this single endpoint.

LM StudioMistral Small 3.2M3 Ultra

RAG & Semantic Search

Documents, notes, and scans are embedded into Qdrant (768-dim vectors). AnythingLLM provides a conversational interface for querying the full knowledge base.

QdrantAnythingLLM768-dim vectors

Speech-to-Text

WhisperX handles transcription with speaker diarization for the Omi wearable and Home Assistant voice satellites. All processing stays local on Apple Silicon.

WhisperXApple SiliconSpeaker Diarization
24Services
80+Workflows
3Servers
0Cloud Dependencies