Blog
Thoughts on data engineering, HR analytics, and building things that work.
2026-05-21
AiSOC — Self-Host Your Own AI-Powered Security Operations Center (Complete Docker Guide)
AiSOC is the open-source, MIT-licensed AI Security Operations Center that you self-host in minutes. LangGraph multi-agent investigation, 52 built-in connectors, public eval harness, and a full SOC console — all in one Docker stack.
2026-05-20
Homelable — Self-Hosted Infrastructure Visualizer for Your Homelab
Homelable is an open-source, self-hosted infrastructure visualizer that auto-discovers your network devices, maps Zigbee topology, monitors service health, and renders everything as an interactive canvas diagram. Complete Docker setup guide.
2026-05-19
Ratty — The GPU-Powered Terminal Emulator with Inline 3D Graphics That's Taking the Command Line Into the Third Dimension
Self-host Ratty — a GPU-rendered terminal emulator built in Rust with Ratatui and Bevy that brings inline 3D objects, a 3D mode, and spinning 3D cursors to your command line. Featured on The Register and It's FOSS.
2026-05-18
Mirage: A Unified Virtual Filesystem for AI Agents — Complete Guide
Self-host Mirage, the open-source virtual filesystem that unifies S3, Gmail, Slack, GitHub, and Redis under a single bash-compatible tree for AI agents. Full CLI, Python, and TypeScript setup guide.
2026-05-17
HTML Anything — The Agentic HTML Editor That Lets Your AI Ship Directly
Learn how to use HTML Anything from nexu-io — a local-first, zero-API-key HTML editor that lets your AI coding agent (Claude Code, Codex, Gemini CLI, Cursor Agent, and 5 more) turn Markdown, CSV, or JSON into ship-ready HTML in seconds.
2026-05-16
Zero-Native: Build Native Desktop Apps with Zig and Web UI (Vercel Labs)
Learn how to use zero-native by Vercel Labs — a Zig-powered framework that builds tiny, fast native desktop apps using your favorite web frontend framework.
2026-05-15
Run DeepSeek V4 Flash Locally with ds4.c — antirez's Purpose-Built Inference Engine
Learn how to run DeepSeek V4 Flash on your own machine using ds4.c — a fast, purpose-built inference engine written in C by Redis creator antirez. Step-by-step setup for macOS Metal and Linux CUDA.
2026-05-14
CubeSandbox — Instant, Secure KVM Sandboxes for AI Agents (Self-Hosted Guide)
CubeSandbox is a Rust-powered, KVM-based secure sandbox service that launches isolated MicroVMs for AI agents in under 60ms with less than 5 MB overhead per instance. Complete self-hosting tutorial with PVM deployment.
2026-05-13
Open Design — The Open-Source Alternative to Claude Design
Open Design is a local-first, open-source alternative to Anthropic's Claude Design. Self-host it with Docker, connect any AI coding agent, and generate production-ready UI prototypes, pitch decks, and marketing assets — all with your own API keys and no cloud lock-in.
2026-05-08
SnapOtter — Self-Host Your Own Image Editor with 48 AI-Powered Tools in One Docker Container
SnapOtter is a self-hosted image manipulation platform with 48 tools (resize, crop, compress, background removal, upscaling, OCR) and local AI, all in a single Docker container. Complete setup tutorial with Docker Compose.
2026-05-07
Glance — The Ultimate Self-Hosted Dashboard for All Your Feeds in One Place
Glance is a lightweight, self-hosted dashboard that aggregates RSS feeds, Hacker News, Reddit, weather, YouTube, GitHub releases, and more in a single beautiful interface. Complete Docker setup tutorial with configuration examples.
2026-05-06
Self-Host Your Docker Infrastructure with Arcane — A Modern Docker Management UI
Learn how to deploy Arcane, a modern open-source Docker management UI with a clean SvelteKit frontend and Go backend — perfect for homelabs and self-hosted infrastructure.
2026-05-05
Slash LLM Token Costs by 80% with RTK — The Rust CLI Proxy Every AI Developer Needs
RTK is a high-performance Rust CLI proxy that sits between your AI coding agent and the shell, filtering and compressing command output to reduce LLM token consumption by 60-90%. Save money, speed up workflows, and keep your context window clean.
2026-05-04
Lightpanda: The Headless Browser Built from Scratch for AI Agents
Lightpanda is a headless browser written from scratch in Zig — not a Chromium fork. 9x faster and 16x less memory than Headless Chrome. Self-host it with Docker for AI agent browser automation, web scraping, and MCP integration.
2026-05-03
Self-Host Opik: Open-Source LLM Observability with Docker
Deploy Comet Opik — the open-source LLM observability platform — with Docker Compose. Trace, evaluate, and monitor your AI applications from prototype to production with 40+ framework integrations.
2026-05-02
MemPalace — Local-First AI Memory System That Achieves 96.6% Retrieval Recall Without Any LLM
Self-host MemPalace, the open-source AI memory system that stores verbatim conversation history with 96.6% R@5 retrieval on LongMemEval — no API key, no cloud, no LLM required. Complete setup guide with MCP integration.
2026-05-02
Self-Host Your Travel Planner with TREK: Real-Time Collaboration, Maps & MCP
Set up TREK, an open-source self-hosted travel planner with real-time collaboration, interactive maps, budget tracking, packing lists, and a built-in MCP server for AI assistants — all in one Docker container.
2026-05-02
Obscura — A Rust-Powered Headless Browser That's 10x Faster Than Chrome for AI Agents
Meet Obscura, the open-source headless browser built in Rust. 30 MB RAM, 85 ms page loads, real V8 JavaScript, CDP-compatible with Puppeteer and Playwright. Complete setup tutorial.
2026-05-02
Monitor Your NVIDIA GPUs in Real Time with GPU Hot
Set up a self-hosted, real-time GPU monitoring dashboard in under 60 seconds with GPU Hot — a lightweight Docker-based tool that tracks utilization, temperature, power draw, and more across single or multi-node clusters.
2026-05-01
Bytebot: Self-Host Your Own AI Desktop Agent (Complete Docker Guide)
Learn how to deploy Bytebot — an open-source AI desktop agent that has its own virtual computer to automate tasks for you.
2026-05-01
Browser Harness: Connect Your LLM Directly to a Real Browser (Self-Hosted Guide)
Learn how to set up Browser Harness — an open-source self-healing CDP harness that connects LLMs directly to your real browser for task automation.
2026-05-01
Self-Host OpenHands v1.7.0: Your Open-Source AI Coding Agent (Step-by-Step Guide)
OpenHands just released v1.7.0 — a major update to the leading open-source AI coding platform. Here's how to self-host it on any Linux VPS with Docker in under 10 minutes.
2026-05-01
Hermes Agent on Dokploy: Build a Self-Hosted AI Development Agent with MCP
A technical guide to deploying Hermes Agent on Dokploy using Docker Compose and wiring it to Dokploy's own MCP server — creating a powerful, persistent AI development agent on a budget VPS.
2026-05-01
Welcome to My Blog
A quick introduction to this space — what I write about and why.