Deploy a full-stack app with PostgreSQL, MySQL, MongoDB, or Redis and a background worker in one command, with DATABASE_URL injected. No Dockerfile needed.
What one deployment includes
On the NEXUS AI runtime, a deployment is your app plus the services it needs on a private network: PostgreSQL, MySQL, MongoDB, or Redis, a background worker running your own command, persistent volumes, and S3-compatible buckets. The app receives DATABASE_URL, REDIS_URL, or MONGO_URI automatically, and the deployment comes with a public HTTPS URL, logs, health checks, backups, and scaling.
The NEXUS AI runtime or your cloud
Deploy to Google Cloud Run, AWS App Runner, or Azure Container Apps and the app runs as a single container, with PostgreSQL or MySQL provisioned as a managed database (Cloud SQL, RDS, or Azure Database) and its connection injected. Redis, MongoDB, and background workers run only on the NEXUS AI runtime; on a cloud target, point the app at an external Redis or MongoDB through environment variables.
No Dockerfile needed
NEXUS AI detects Node.js, Next.js, React, Vue, Angular, Vite, Python, FastAPI, Django, PHP, Laravel, Symfony, Ruby on Rails, and static sites and generates the build. A valid Dockerfile at the root of the app folder is used instead when there is one, and --build-command, --start-command, and --install-command override the detected steps.
Connected on the first deploy
Database containers start first, and the app and worker wait until they report healthy, so the first deploy connects instead of crashing and retrying. A managed cloud database created at deploy time is ready before the app is wired to it. Point your ORM at DATABASE_URL and run migrations in the start command, for example prisma migrate deploy && npm start.
Background workers on the same network
Add --worker-command to run a queue consumer or scheduler, such as a BullMQ, Celery, or Sidekiq worker. The worker runs from the same image as the app with the same environment variables, so it reaches the same database and Redis without extra configuration.
Repos with a separate frontend and API
When a repository keeps frontend/ and backend/ side by side, name the folder to deploy with --root-dir or the Root Directory field: deploy the API with its database, then deploy the frontend with the API URL in an environment variable. Frameworks that serve pages and API from one app, such as Next.js, Django, Rails, and Laravel, deploy as a single deployment.
Deploy a full-stack app, step by step
Next.js with Postgres, Redis, and a worker
One command builds the app, starts PostgreSQL and Redis, and runs a queue worker next to it.
Install the CLI and sign in.
Deploy the repository with the databases and worker command it needs.
Upload a ZIP or folder, or use the project repository, and pick services in the form.
Open a project and choose Deploy Full App.
Choose the source, set Root Directory if the repo holds several apps, and add PostgreSQL, MySQL, MongoDB, Redis, or a worker under Additional Services.
Deploy, then follow status, logs, and the public URL on the Deployments page.
Project → Deploy Full App → Source: repository, ZIP, or folder
Root Directory: backend
Additional Services: PostgreSQL, Redis
Deploy
Build and start commands are optional; leave them empty to use the detected ones.
Choose the source: a Git repository, a ZIP, or a folder, and set the root directory if the repo holds several apps.
Add the databases and worker the app needs.
Deploy, then follow logs, the public URL, backups, scaling, and rollback.
What a full-stack deployment supports on each target
NEXUS AI runtime
Google Cloud Run, AWS App Runner, Azure Container Apps
App
Container with a public HTTPS URL
Single container; in your own cloud account when you connect one
PostgreSQL and MySQL
Run inside the deployment
Managed Cloud SQL, RDS, or Azure Database
Redis and MongoDB
Run inside the deployment
Not provisioned; connect an external service with environment variables
Background workers
Worker with your command, same image and environment
Not supported; use the NEXUS AI runtime
Connection variables
DATABASE_URL, REDIS_URL, MONGO_URI
DATABASE_URL and the host, port, user, password, and database name
Plans
Every plan
Google Cloud on Starter; AWS, Google Cloud, and Azure on Pro
Detected automatically
Framework
Detected from
Deploys as
Next.js
package.json
One app serving pages and API routes
Express, NestJS, and other Node.js servers
package.json
One app; pair with a frontend deployment if the UI is separate
React, Vue, Angular, Vite
package.json with an index.html entry
Static frontend
Django, FastAPI, Flask
manage.py or requirements.txt
One Python app
Laravel, Symfony, PHP
composer.json
One PHP app
Ruby on Rails
Gemfile
One Ruby app
Frequently asked questions
What is the easiest way to deploy a full-stack app with a database?
Run nexus deploy source with the repository URL and --services for the databases you need, or use Deploy Full App in the dashboard. NEXUS AI builds the app without a Dockerfile, starts the databases, injects DATABASE_URL, and returns a public HTTPS URL.
How do I deploy a frontend and backend from one repository?
Deploy each folder as its own deployment with --root-dir (or the Root Directory field in the dashboard): the API with its database first, then the frontend with the API URL in an environment variable. Without a root directory, automatic detection deploys only one of the two folders.
Do I need a Dockerfile?
No. NEXUS AI detects Node.js, Next.js, React, Vue, Angular, Vite, Python, Django, FastAPI, PHP, Laravel, Symfony, Rails, and static sites. If the app folder has a valid Dockerfile, it is used instead.
Which databases can I use on each target?
On the NEXUS AI runtime: PostgreSQL, MySQL, MongoDB, and Redis inside the deployment. On Google Cloud Run, AWS App Runner, and Azure Container Apps: managed PostgreSQL or MySQL. Redis and MongoDB are not provisioned on cloud targets; connect an external service instead.
Can workers connect to the same database and Redis services?
Yes. A worker started with --worker-command runs from the same image with the same environment variables, so it reaches the same Postgres, Redis, and buckets. Workers run on the NEXUS AI runtime.
How do database migrations run?
Put them in the start command, for example npx prisma migrate deploy && npm start or python manage.py migrate && gunicorn app.wsgi. The app starts only after the database reports healthy, so migrations run against a ready database.
Can I restore data from one deployment database to another?
Yes. Back up a deployment database, then restore it in place or into another database of the same engine in your organization, for example to refresh staging from production.
Is NEXUS AI a Heroku alternative for full-stack apps?
For apps that need a web process, a database, and a worker together, yes. On Heroku, dynos and add-ons such as Postgres and Redis are priced separately. On NEXUS AI they deploy together, and the NEXUS AI runtime is included in a flat monthly plan.
What does it cost?
The Free plan includes one deployment for testing. Starter is $29 a month with two deployments, a custom domain, and Google Cloud. Pro is $149 a month with five deployments, up to 10 concurrent containers, rollback, and AWS, Google Cloud, and Azure.