Скачать или смотреть Deploy ANY AI App to the Cloud FAST

It's fun to build your AI apps locally, but in the end when you want to use it for real with other users on your platform, it can't be stuck in your development environment. Simply put, it has to be made available over the internet for other people to access in a secure way.

So in this video, I walk you through step by step how to deploy your AI apps to the cloud, using the local AI starter kit by n8n as the example! But this guide will work for really ANY AI app - built with Streamlit, OpenWebUI, Next.js, Docker apps, etc.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

00:00 - Plan of Attack
01:39 - What I'm Deploying as an Example
02:52 - Cloud Machine Setup w/ DigitalOcean
05:59 - Steps 1-2 - Clone repo and cd into it
06:10 - Steps 3-4 - Edit the .env and docker-compose.yml files
08:08 - Steps 5-7 - Firewall Setup
09:08 - Steps 8-13 - Reverse Proxy Setup with Nginx
12:51 - Step 14 - DNS Record Setup
14:06 - Steps 15-16 - SSL Certificate Installation
15:02 - Step 17 - Spin up the Containers!
16:57 - Testing our AI App Deployment
20:49 - Outro

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Below are the EXACT steps I cover in the video. This will work on any cloud provider as long as you have a Ubuntu Linux machine. If you don't, some of the steps might have to be a bit different but the process will be very similar.

Most of the steps will be the same regardless of what you are deploying! Steps 1-4 and 17 are specifically for the local AI starter kit, the rest are very app agnostic.

Prerequisite: You have a machine in the cloud with Docker installed.

STEPS:

1. git clone https://github.com/n8n-io/self-hosted...

2. cd self-hosted-ai-starter-kit

3. Edit the .env file to edit POSTGRES_USER, POSTGRES_PASSWORD, POSTGRES_DB, N8N_ENCRYPTION_KEY, and N8N_USER_MANAGEMENT_JWT_SECRET

You can edit the .env by entering the command "nano .env", adding in the contents, and then saving with "Ctrl/Cmd + X, then enter Y, then hit enter".

4. Edit the docker-compose.yml file to set the Ollama command on line 44 to:

"sleep 3; OLLAMA_HOST=ollama:11434 ollama pull llama3.1; OLLAMA_HOST=ollama:11434 ollama pull nomic-embed-text"

5. sudo ufw enable

6. sudo ufw allow 80 && sudo ufw allow 443

7. sudo ufw reload

8. sudo apt install nginx

9. sudo nano /etc/nginx/sites-available/local-ai-app

10. Put in the following content (with nano just like with the .env file):

server {
listen 80;
server_name [your subdomain].[your domain].com;
location / {
proxy_pass http://localhost:5678;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection 'upgrade';
proxy_set_header Host $host;
proxy_cache_bypass $http_upgrade;
}
}

Change localhost:5678 to the port your app runs on if different!

11. sudo ln -s /etc/nginx/sites-available/local-ai-app /etc/nginx/sites-enabled/

12. Run "sudo nginx -t" just to verify the config

13. sudo systemctl reload nginx

14. Set DNS record for a subdomain to point to IPv4 of cloud machine (A record, host is your subdomain you want to use, value is the IPv4 of the machine)

15. sudo apt install certbot python3-certbot-nginx

16. sudo certbot --nginx -d [your subdomain].[your domain].com

17. docker compose --profile cpu up -d (can run with Nvidia GPU enabled too)

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Here is a link to the Local AI starter kit by n8n!

https://github.com/n8n-io/self-hosted...

If you want to host on DigitalOcean like me, here is the link for that as well:

https://www.digitalocean.com/

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Artificial Intelligence is no doubt the future of not just software development but the whole world. And I'm on a mission to master it - focusing first on mastering AI Agents.

Join me as I push the limits of what is possible with AI. I'll be uploading videos at least two times a week - Sundays and Wednesdays at 7:00 PM CDT! Sundays and Wednesdays are for everything AI, focusing on providing insane and practical educational value. I will also post sometimes on Fridays at 7:00 PM CDT - specifically for platform showcases - sometimes sponsored, always creative in approach!