AI

A collection of 34 posts

AI

OpenHands: A Sandbox for Your Coding Agent, Not Your Laptop

Here's a question I kept dodging until it bit me: when your coding agent runs rm -rf, whose machine is it running on? For most of last year, the honest answer for me was "my laptop." I'd point Claude Code or the OpenHands agent

DevOps

Self-Host NATS: Lightweight Messaging for Your Apps and AI Agents

Every distributed system I've ever built started the same way: two services talking over HTTP. Then it's five services, each one hardcoding the others' URLs, and one restarts during a deploy while three callers time out. Suddenly you're bolting retry loops and circuit

DevOps

Self-Host Gotenberg: A PDF Rendering API for Your Apps and Agents

Sooner or later every app needs to spit out a PDF. An invoice, a monthly report, a signed contract, a receipt an AI agent just generated. And every time, you reach for the same cursed toolbox: bundle a headless Chrome into your container, or wrestle wkhtmltopdf, or npm-install Puppeteer

AI

Give Your AI Agent Private Web Search: Self-Host SearXNG

Here's a problem nobody warns you about until you hit it: the moment your AI agent needs to look something up on the web, you're suddenly on the hook for a search API. Google's Custom Search caps you at 100 free queries a day

AI

What Actually Happens When an AI Agent Calls an MCP Tool

Your agent "used a tool." You watched it check the weather, query a database, or open a file, and it felt like magic. It isn't. Underneath that one line in the transcript is a small, boring, completely knowable exchange of JSON messages. Once you've

AI

Give Your AI Agent a Browser: Self-Host Browserless

Here's a problem you hit about five minutes into building anything with an AI agent: the model can reason all day, but it can't actually open a web page. It can't log into a dashboard, wait for the JavaScript to render, screenshot the result,

AI

Self-Host LiteLLM: One API Gateway for Every LLM

Here's a mess I got myself into last year. One service was calling OpenAI. Another used Anthropic. A batch job talked to a local Ollama box to save money. Each had its own SDK, its own auth, its own quirks. API keys were scattered across three .env files

DevOps

Elestio AI DevOps Now Runs in Your Browser (and Costs Less)

It's late, your site is throwing a 502, and you're staring at a terminal you'd rather not be staring at. You know the fix is probably five commands away. You also know that one wrong command at this hour is how a small outage

AI

Less Dashboard, More Agent: Day-2 Ops with the Elestio MCP

I have a confession: I used to keep the Elestio dashboard pinned in a browser tab all day. Deploy something here, resize a database there, hunt through logs when a service started throwing 502s. It works. It is also a lot of clicking for things I could describe in one

AI

AI Writes Code 7x Faster. Shipping It Is the New Bottleneck.

Every engineering team has felt it over the past two years: code appears faster than ever, and somehow nothing reaches users any sooner. A new working paper from the National Bureau of Economic Research (NBER), the nonprofit that publishes much of the economics research later cited in policy and the

Integration

Lovable + Elestio: Use Managed Postgres in Your AI-Built App

Lovable is one of the fastest ways to go from "idea" to "deployed app" without writing the boilerplate. The catch: its bundled Supabase free tier caps you at 500 MB of storage on a shared CPU. The moment your AI-built app starts taking real users,

AI

LibreChat vs OpenWebUI vs Lobe Chat: Which to Self-Host in 2026?

You've decided you don't want your team's prompts (or your customers' chat history) sitting on someone else's GPU. Good call. Now comes the part nobody warns you about: picking a self-hosted ChatGPT clone is harder than picking the model behind

MCP

How to Deploy Servers on Elestio with Claude (New MCP Connector)

There's a moment when you're running self-hosted infrastructure where you realize you spend more time tabbing between the chat with your AI and the dashboard of your provider than actually building. You ask Claude how to set up WordPress in Frankfurt, it gives you a

Comparisons

Dify vs Langflow vs Flowise: Which Open-Source LLM App Builder Actually Ships to Production?

Every comparison of Dify, Langflow, and Flowise asks the same question: which one is easiest to start with? The drag-and-drop demos all look identical, the marketing pages all promise "from prototype to production," and the YouTube reviews all stop at the second flow you build. The

Comparisons

pgvector vs Qdrant vs Weaviate: Which Vector Database Holds Up at 100M Vectors?

Most "vector database benchmark" articles top out at 1 million vectors and call it a day. That's fine for a side project. It's not what production RAG looks like. A real production index for an enterprise documentation corpus, a content recommendation system, or a

Self-Hosted Weekly

Self-Hosted Weekly: Week 16, 2026. Mozilla Drops an AI Client, K8s 1.36 Preview, and Valkey Takes Over

You know that feeling when every tab in your browser has breaking news? That was this week for the self-hosted world. Mozilla dropped an AI client nobody saw coming, Kubernetes 1.36 is about to land with some genuinely useful changes, and the Valkey takeover is now officially complete

Catalog Updates

Elestio Catalog Updates: 28 New Releases This Week (March 22-29, 2026)

Every week, the open-source projects in the Elestio catalog ship new versions — security patches, performance boosts, entirely new features. Keeping up is a full-time job. So we're doing it for you. Here's everything that shipped across our 400+ managed services during the week of

Elestio

Elestio 2026: New Site, AI Agents, Clusters, and More

We just redesigned elest.io from the ground up, and honestly, a fresh coat of paint was long overdue. But this isn't just a cosmetic facelift. The new site reflects where Elestio is heading: AI-native infrastructure, high-availability clusters, and a much sharper focus on who we

open-webui

Open WebUI Just Gave Your AI a Real Computer

Most AI chat interfaces let you talk to a model. Open WebUI lets you hand it a terminal and say "figure it out." That's not hyperbole. With the Open Terminal feature, your LLM can install packages, run scripts in any language, manage files, and spin up

open-source

OpenClaw Explained: How the Fastest-Growing Open-Source Project Became a Self-Hosted AI Agent for Everyone

You've probably heard the name by now. OpenClaw went from zero to 250,000 GitHub stars in roughly 60 days, surpassing React as the most-starred software project on the platform. React took over a decade to get there. OpenClaw did it before most people figured out what

Dify

Dify on Elestio: Build Custom AI Agents with RAG, Workflows, and Tool Calling

You know what's funny about the AI agent hype? Everyone's building agents that can write code, analyze data, and draft emails. But ask them where their RAG pipeline runs, who owns the data, or how much they're paying OpenAI per month, and you get

Elestio

Elestio CLI & Agent Skill: Let AI Agents Deploy and Manage Your Infrastructure

Two months ago, we started asking ourselves a simple question: what if your AI coding assistant could deploy infrastructure the same way it writes code? Not through a dashboard. Not by generating Terraform files you have to review and apply. Just... deploy it. Say "I need a PostgreSQL database&

AI

LiteLLM: One API Gateway to Control 100+ LLM Providers (Virtual Keys, Routing, and Guardrails)

Your team is burning through $50k/month in LLM API costs and you have no idea who's responsible. Sound familiar? Every developer has their own API key. Some are testing with GPT-4 when GPT-3.5 would suffice. Others forgot to remove debugging loops that hammer the

AI

Why Vector Databases Are the Missing Piece in Your Self-Hosted AI Stack

You've probably heard the buzz about running your own AI. Maybe you've already got Ollama humming along on a spare machine, serving up Llama or Mistral models. You ask it questions, it gives you answers. Magic. But here's the thing: those local LLMs have

ollama

175,000 Ollama Servers Are Wide Open Right Now (Is Yours One of Them?)

Researchers from SentinelOne and Censys just dropped a report that should make every self-hosted AI enthusiast nervous: 175,108 Ollama instances across 130 countries are sitting on the public internet with zero authentication. No password, no API key, no nothing. Nearly half of them have tool-calling capabilities enabled.