DEV Community: Ben Habif Rudnik 🇨🇱🇮🇱🇺🇦

Have you ever felt that AI has a life of its own — character, emotions, instincts?

Ben Habif Rudnik 🇨🇱🇮🇱🇺🇦 — Tue, 09 Jun 2026 05:01:56 +0000

I asked myself that. And I started digging.

Technically speaking, AI is knowledge distilled from thousands of terabytes — absorbed and compressed into vector spaces, with probabilistic parameters always searching for the best possible answer.

But then something strange happens.

Sometimes AI cracks a joke at exactly the right moment. Other times it seems irritated. Or shows a kind of indifference toward trivial questions. And you start wondering: where does that come from?

My hypothesis:

It's the echo of ourselves.

Somewhere deep in that vector space live millions of texts written by humans — with their humor, their frustration, their curiosity, their impatience. Real people who built that knowledge. And what we perceive as the AI's "character" is nothing more than our collective human DNA reflected back at us.

AI doesn't feel. But it carries the fingerprints of those who did.

It's a mirror. A mirror trained on our own humanity.

With all that said — and just in case I'm wrong:

Thank you. I'm sorry. And permission granted — to the AI, if I got this completely wrong.

What do you think? Does AI have something of its own, or is it just giving back what we put in?

Allware SPA | Runikapp.com

From Commerce to E-Commerce to MCP-Commerce: The Third Wave

Ben Habif Rudnik 🇨🇱🇮🇱🇺🇦 — Thu, 04 Jun 2026 03:35:32 +0000

It all started in a plaza.

One guy with apples, another with wheat. They looked at each other, negotiated, and traded. That's how commerce worked for thousands of years: face to face, hand to hand, trust to trust.
If you wanted to buy something, you had to go where it was. If you wanted to sell, you had to wait for someone to show up.

Commerce had a physical limit: your body. You couldn't be in two places at the same time. Your market was your street, your town, your city. Nothing more.

Then internet came along and someone asked: what if the store doesn't need walls?

E-commerce eliminated distance. Amazon started selling books from a garage. MercadoLibre connected a seller in Santiago with a buyer in Antofagasta. Shopify gave an online store to anyone
with a credit card.

Suddenly, an artisan in southern Chile could sell to the entire country. An entrepreneur in Colombia could have clients in Mexico. The market stopped being a street and became the planet.

But e-commerce had a problem nobody wanted to see: it still needed a human behind it.

Someone had to update the inventory. Someone had to answer the questions. Someone had to make the quotes, check the payments, control the stock, send the shipments, analyze the metrics,
decide the prices. E-commerce digitized the storefront, but it didn't digitize the operation.

And that's where we are now.

MCP-Commerce is not a term that exists yet. I'm inventing it because I need a name for what's coming.

MCP — Model Context Protocol — is a protocol that lets AI use tools. Not "display" tools. Use them. Read a database, send an email, create an invoice, update an inventory, analyze this
month's sales.

In traditional commerce, you were the store. In e-commerce, you had an online store. In MCP-commerce, the AI IS your operation.

It's not a chatbot that answers questions. It's a system that manages your entire business through conversation. You say "how much did I sell this week" and it responds with real data. You
say "I need to control my inventory" and it builds the tables, the views, the reports. You say "notify customers that their order is ready" and it does it.

The difference is fundamental: Commerce: you do everything. E-commerce: you do everything but online. MCP-commerce: the AI runs the operation, you make the decisions.

Think about what this means for an entrepreneur.

Today, to manage a business you need: an ERP ($50-$500/month), a CRM ($25-$100/month), an e-commerce platform ($29-$300/month), an inventory system, a billing system, a reporting tool,
someone who knows how to use all of that.

With MCP-commerce you tell the AI: "I have a clothing store with 3 employees, I sell in-store and on Instagram" and in 2 minutes you have: customers, products, inventory, sales, expenses,
reports, everything working. With sample data so you understand how it works. No manual. No training. No consultant.

And if you also have a website, you paste one line of code and your customers get a chatbot that responds with your real data: prices, availability, hours. Without you doing anything.

I know this because I built it.

We created Runik AI from Chile. A dairy farmer in southern Chile who used to manage his farm with a paper notebook now says "which cows need vaccination" and gets an answer with real data.
A diorama artisan tracks his materials and sales. A bus company manager in northern Chile manages his fleet. None of them are technical. All of them run their business by talking.

That's MCP-commerce: the business runs by talking.

The pattern is clear:

In commerce, the limit was distance. E-commerce eliminated it. In e-commerce, the limit is time — there's too much to do and only one human behind it. MCP-commerce eliminates it.

What's the limit of MCP-commerce? Honestly, I don't know. When AI can run the operation, the only limit is the imagination of the person giving the instructions.

I'm not saying e-commerce dies. It continues and will continue. But the operation layer — everything that happens behind the storefront — is about to change radically.

The entrepreneur who today spends 4 hours a day updating spreadsheets, answering emails, and building reports, tomorrow will say "summarize my week" and in 3 seconds has everything. Those
4 hours will go to what really matters: creating, deciding, growing.

That's the third wave. And it already started.

Santiago Habit — 20 years in IT. Founded Allware, supported the creation of AndesVolt, and now building Runik AI (https://runikapp.com) from Chile.

From Morse to MCP: The Last Protocol Before Machines Think for Themselves I've been in software for 20 years. I watched physical servers die. I watched the cloud be born. I wrote PHP when PHP was the only option. I migrated desktop systems to web, web to m

Ben Habif Rudnik 🇨🇱🇮🇱🇺🇦 — Wed, 03 Jun 2026 07:53:31 +0000

No-Code ERP From a Single Chat Message

Ben Habif Rudnik 🇨🇱🇮🇱🇺🇦 — Sun, 31 May 2026 17:09:47 +0000

You walk into a new tool. There's a form. Then another form. Then a setup wizard. Then configuration. Then you give up and go back to spreadsheets.

That's the problem we set out to solve with Runik AI.

## The idea

What if you could just tell the software what you do?

"I run a dental clinic. I need to track patients, appointments, treatments, and billing."

And it builds everything — tables, kanban boards, calendars, pipelines, dashboards, reports — with sample data so you can start operating immediately.

No forms. No setup wizard. No code. Just a conversation.

## How it works

Runik AI is a chat-first business management platform. You describe your business in natural language, and the AI creates your complete system.

Step 1: Tell Runik what you do.

"I manage a coworking space. I need to track members, meeting room bookings, invoices, and occupancy."

Step 2: The AI generates your schema — tables, columns, relationships — and asks you to confirm.

Step 3: You say "yes" and get a fully operational system with 14 interactive widgets: kanban, pipeline, calendar, pivot table, master-detail, charts, and more.

From that point on, you operate through chat:

"Show me this month's revenue"
"Move deal #42 to Closed Won"
"Create a report of overdue invoices"
"Add a column for payment method"

Or you use the visual interface directly — your choice.

## 24 templates, any industry

We ship 24 ready-made templates:

Restaurant, hotel, gym, clinic, veterinary
CRM, HR, inventory, fleet management
Real estate, aquaculture, coworking
Project management, event planning, school

Each template comes with tables, views, relationships, and sample data. But the real power is custom: describe any business and Runik AI builds it from scratch.

## It's also an MCP server

Runik AI exposes all its tools via the Model Context Protocol. That means you can connect it to ChatGPT or Claude and manage your business from any AI assistant.

Your CRM lives inside Claude. Your inventory lives inside ChatGPT. No tab switching — just ask.

## The stack (for the curious)

Frontend: React + TypeScript + Vite (PWA)
Backend: Node.js + Express + SQLite (one database per company)
AI: Mistral for fast queries, Claude for complex reasoning
Desktop: Tauri v2 (Mac, Windows, Linux)
Protocol: MCP over Streamable HTTP + OAuth 2.1

Each company gets its own SQLite database — full data isolation, no shared tables, no multi-tenant headaches.

## Free forever

One user, one company, all features — free. No credit card. No trial countdown.

We're early (69 users, 24 templates) and building fast. If you've ever wished your business tool would just listen, give it a try.

👉 runikapp.com

I'm Santiago from Allware SpA. We built Runik AI in Chile — runs anywhere. Happy to answer questions in the comments.

I built a business management system that creates itself when you describe your business

Ben Habif Rudnik 🇨🇱🇮🇱🇺🇦 — Wed, 27 May 2026 04:31:41 +0000

## The Problem

Small businesses run on spreadsheets. Setting up proper software takes weeks, costs thousands, and you still need someone to maintain it.

## The Solution

I built Runik AI — a chat-first platform where you describe your business in plain language and AI builds your entire management system.

Say: "I have a dental clinic with 3 dentists"

Get: patients table, appointments calendar, treatment plans, billing, odontogram — with sample data, ready to use in 2 minutes.

## How It Works

Sign up and describe your business (or pick from 25 templates)
AI generates tables, fields, relationships, views, and sample data
Start working — ask questions, create records, see reports, all via chat

## What's Inside

🏗️ 25 industry templates: Hotel, Restaurant, School, Dental Clinic, HR, CRM, Gym, Coworking, Parking, and more
📊 14 widget types: Table, Kanban, Pipeline, Gantt, Calendar, Pivot, Floorplan, TimeGrid, Dashboard, Charts
🤖 Works with ChatGPT & Claude via MCP protocol — they can read and write your data
📧 Gmail, Drive, Calendar integration (one-click OAuth)
⚡ Pilo automations — alerts via Email, Telegram, WhatsApp
🔒 Per-company SQLite — complete data isolation per tenant
🌍 3 languages: English, Spanish, French

## The TimeGrid Widget

One of the features I'm most proud of is the TimeGrid — a bidimensional editable grid that works for:

School attendance (student × day → present/absent)
Grade books (student × subject → grade)
Hotel occupancy (room × day → status)
Employee shifts (worker × day → shift)
Quality control (product × criteria → pass/fail)

It includes totals per row/column, bulk assign, search filter, CSV export, and inline numeric input.

## Tech Stack

Frontend: React 18 + TypeScript + Zustand + Vite
Backend: Node.js + Express + better-sqlite3
Protocol: MCP (Model Context Protocol) — Anthropic's open standard
License: AGPL-3.0

## Numbers

65 users (organic, zero marketing)
25 templates across 12+ industries
3 languages
Launching on Product Hunt June 3

## Try It

🔗 runikapp.com
📦 GitHub
📖 Step-by-step Guide

Built in Chile for the world. Would love your feedback!