Set Up Your Own AI Assistant in 30 Minutes

What You Will Build

By the end of this guide, you will have a personal AI assistant running on dedicated infrastructure, connected to your messaging app of choice, ready to help with research, writing, analysis, and more. No server setup, no command line, no Docker files. Just sign up, configure, and start chatting.

We will walk through the entire process using Molty by Finna, the managed hosting platform for Moltbot. The whole thing takes about 30 minutes, with most of that time spent waiting for infrastructure to spin up.

What You Need Before Starting

An email address for your Molty account
A phone with WhatsApp installed (or a Telegram account, if you prefer that channel)
Your own AI model API key (from Anthropic, OpenAI, or another supported provider)
A modern web browser

That is it. No technical background required.

Step 1: Create Your Molty Account (2 minutes)

Head to the Molty by Finna website and click "Get Started." You will be asked to create an account using your email or a social login (Google, GitHub, etc.).

Once signed in, you land on the dashboard. This is your control center for managing AI assistants. On a new account, it will be empty - that is about to change.

Every new account starts with a 3-day free trial that includes full access to all features. No credit card required to start.

Step 2: Create Your Gateway (3 minutes)

In Molty terminology, a "gateway" is your AI assistant's home - a dedicated virtual machine that runs your Moltbot instance. Each gateway is isolated from every other user, running on its own VM with its own storage.

Click "Create Gateway" and you will be asked for two things:

Name Your Gateway

Pick something meaningful to you. This is just for your reference in the dashboard - your contacts will not see it. Something like "Personal Assistant" or "Work Bot" works fine.

Choose a Region

Select the geographic region closest to you or your primary audience. Options typically include regions in North America, Europe, and Asia-Pacific. Closer regions mean lower latency in message processing.

Click "Create" and the provisioning process begins. Behind the scenes, Molty is:

Creating a dedicated virtual machine for your gateway
Allocating encrypted storage for your assistant's data
Setting up secure networking
Installing and configuring Moltbot

This takes about 60-90 seconds. You will see a progress indicator on the dashboard.

Step 3: Configure Your AI Model (5 minutes)

Once your gateway is running, you need to tell it which AI model to use. Moltbot supports multiple providers:

Anthropic (Claude models) - recommended for general use
OpenAI (GPT models) - widely supported
Other providers via OpenAI-compatible APIs

Adding Your API Key

Navigate to the "Settings" or "API Keys" section of your gateway. Enter your API key from your chosen provider. The key is encrypted before storage and is never visible again after you save it.

Choosing a Model

Select which specific model your assistant should use. For most people, we recommend starting with a capable general-purpose model. You can always change this later.

A note on costs: The AI model API usage is billed directly by your model provider (Anthropic, OpenAI, etc.), not by Molty. Molty provides the infrastructure; the model provider charges for the AI processing. This gives you full control over your AI spend.

Step 4: Connect a Messaging Channel (10 minutes)

This is where it gets exciting. You are going to connect your assistant to a real messaging app. We will cover the two most popular options.

Option A: Connect WhatsApp

WhatsApp is the most popular choice for personal AI assistants because most people already use it daily.

In your gateway dashboard, navigate to the "Channels" tab
Find WhatsApp and click "Enable"
A QR code will appear on your screen
Open WhatsApp on your phone
Go to Settings > Linked Devices > Link a Device
Scan the QR code with your phone

Once scanned, your gateway establishes a connection to WhatsApp. This typically takes 10-30 seconds. You will see the channel status change to "Connected" in the dashboard.

Important: The WhatsApp connection uses the Linked Devices feature, which means your assistant runs alongside your regular WhatsApp - it does not replace it. Your normal WhatsApp usage continues unaffected.

By default, WhatsApp is configured with "pairing" mode, which means new contacts need to be approved before they can interact with your assistant. This prevents strangers from using your AI credits. You can adjust this in the channel settings.

Option B: Connect Telegram

Telegram is an excellent alternative, especially if you want to keep your AI assistant separate from your primary messaging.

Open Telegram and message @BotFather
Send /newbot and follow the prompts to create a bot
BotFather will give you an API token - copy it
In your Molty dashboard, navigate to the Channels tab
Find Telegram and click "Enable"
Paste your bot token in the configuration field
Save and the connection will be established

Your Telegram bot is now live. Find it by searching for the username you chose during creation and start chatting.

Step 5: Send Your First Message (1 minute)

Open your connected messaging app and send a message to your assistant. Try something like:

"Hello! What can you help me with?"
"Summarize the latest news about renewable energy"
"Write a professional email declining a meeting invitation"

Your assistant will respond within a few seconds. The first message might take slightly longer as the system initializes the conversation context.

Congratulations - you now have a personal AI assistant living in your messaging app.

Step 6: Customize Your Assistant (10 minutes)

Now that the basics are working, you can fine-tune how your assistant behaves.

System Prompt

The system prompt defines your assistant's personality, expertise, and boundaries. Navigate to the agent configuration in your dashboard and set a system prompt. For example:

"You are a helpful personal assistant. You communicate in a clear, concise style. You are knowledgeable about technology, business, and productivity. When you are unsure about something, you say so rather than guessing."

You can make this as specific as you want. Some people configure their assistant with their name, time zone, preferred language, and areas of interest.

Tool Permissions

Moltbot assistants can use various tools. In the settings, you can enable or disable:

Web browsing: Let the assistant search the internet and read web pages
Code execution: Allow the assistant to write and run code for analysis tasks
File management: Enable reading and writing files in the assistant's workspace

For a personal assistant, enabling web browsing is highly recommended. It transforms the assistant from a static knowledge base into a dynamic research tool.

Scheduled Tasks

One of Moltbot's most powerful features is scheduled tasks (cron jobs). You can configure your assistant to proactively send you information at set times:

Morning news briefing at 7:00 AM
Daily weather forecast
Weekly summary of saved articles

These are configured in the gateway settings and run automatically on schedule.

What to Expect in Daily Use

Response Times

Most messages get a response within 3-15 seconds, depending on the complexity of the request and which AI model you are using. Tasks that involve web browsing or code execution may take longer.

Conversation Memory

Your assistant remembers the context of your current conversation. It knows what you discussed earlier in the chat and can reference previous messages. Long-term memory across separate conversations depends on the configuration.

Multi-Channel Behavior

If you connect multiple channels (say WhatsApp and Telegram), each channel maintains its own conversation context. A discussion on WhatsApp is separate from one on Telegram.

Availability

Your gateway runs 24/7 on dedicated infrastructure. It does not sleep, it does not need your computer to be on, and it does not depend on a browser tab. Send a message at 3 AM and it responds immediately.

Troubleshooting Common Issues

"My assistant is not responding"

Check the gateway status in your dashboard. If it shows "Running," the issue might be with the channel connection. Try checking the channel status - if WhatsApp shows "Disconnected," you may need to re-scan the QR code.

"WhatsApp QR code is not loading"

Make sure your gateway is fully started (status shows "Running"). It can take up to 90 seconds after creation for the gateway to be ready. Refresh the page and try again.

"Responses seem slow"

Response time depends on the AI model and the complexity of the request. If you are consistently seeing slow responses, try switching to a faster model variant. Also check if you have web browsing enabled - requests that involve searching the web naturally take longer.

"I want to change the AI model"

You can switch models at any time through the dashboard settings. The change takes effect for new messages immediately.

Next Steps

Once you are comfortable with the basics, explore these features:

Connect additional channels - Add Telegram, Discord, or Slack alongside WhatsApp
Set up daily briefings - Configure scheduled tasks for proactive updates
Refine the system prompt - Experiment with different personalities and instructions
Explore tool capabilities - Try asking your assistant to research topics, analyze data, or help with code

Your AI assistant is a tool that gets more valuable the more you use it. As you discover new use cases and refine the configuration, it becomes increasingly tailored to how you work and what you need.

Welcome to the future of personal AI. It lives in your pocket, it knows your preferences, and it is always ready to help.