---
title: Deploy NocoDB with AI on Blossom
url: "/deploy/nocodb/"
canonical_url: https://docs.blossom88.com/deploy/nocodb/
markdown_url: https://docs.blossom88.com/deploy/nocodb.md
summary: Deploy NocoDB with AI on Blossom. Build a spreadsheet-style workspace with
  forms, an authenticated API, managed PostgreSQL and Redis, and private S3 or R2
  attachments.
categories:
- deployment
- business-tools
---

# Deploy NocoDB with AI on Blossom

[NocoDB](https://nocodb.com/) turns database records into a spreadsheet-style
workspace. Use it to track projects, collect information through forms, and
connect your data to other apps through an API.

It's written in TypeScript, with a NestJS backend and a Vue/Nuxt frontend.
This setup uses PostgreSQL. Browse the
[source on GitHub](https://github.com/nocodb/nocodb).

Run NocoDB on regular App Capacity, with managed PostgreSQL and Redis. Store
attachments in your own private S3 or Cloudflare R2 bucket so uploads survive
Container replacement.

[Install the Blossom CLI and skill](/learn/getting-started/install/)
on your agent's machine and choose a Blossom Cloud Location when you log in.
You can also use
[Blossom BYOC (bring your own cloud)](/learn/getting-started/choose-your-cloud/#blossom-byoc).

## Ask Your Agent

> **Ask your AI agent:** Deploy NocoDB on Blossom with R2 for files. Help me build a content tracker with a form and API.

> Follow this guide: https://docs.blossom88.com/deploy/nocodb/



Your agent checks the current
[NocoDB deployment requirements](https://nocodb.com/docs/self-hosting/environment-variables),
chooses compatible maintained releases, and saves the configuration.

| Part | Purpose |
|---|---|
| Web Service | Spreadsheet workspace, forms, and API |
| Worker Service | Background tasks using the same NocoDB image |
| Managed PostgreSQL | Accounts, workspace configuration, and records |
| Managed Redis | Shared background queue |
| Your private bucket | Uploaded attachments |

Use stable `NC_AUTH_JWT_SECRET` and `NC_CONNECTION_ENCRYPT_KEY` values on
both Services. Give both the same database, Redis connection, storage settings,
and canonical HTTPS `NC_SITE_URL`. Start the web Service first so its initial
database setup completes before enabling the worker.

### Connect Your Bucket

Create a private bucket in your own provider account and a bucket-restricted
credential. Supply its endpoint, region, bucket, and credentials through a
private environment file. Keep provider administration credentials off the app.

Enable secure attachments so downloads use signed links. With R2, use
path-style requests. Upload and download a sample file before adding real data.

### Create Your Workspace

Keep NocoDB behind **Blossom Account access** while creating the owner.
Open the generated HTTPS address, finish setup, and create a base. Add a table
with a title, status, notes, and attachment field.

Create a form view to collect a new record. The shared form still follows your
Blossom access policy; explicitly configure public access if people outside
your Account need to use it. Configure email before relying on invitations or
password resets.

## Put Your Tracker to Work

> **Ask your AI agent:** Add a few sample articles, submit one through the form, and help me read the records with an API key.

> Follow this guide: https://docs.blossom88.com/deploy/nocodb/



Create an API token scoped to the base and keep it private. Check that an
unauthenticated records request is refused. Submit your form and confirm the
new record appears in the table and API response.

![A NocoDB content tracker with article statuses, notes, and private file attachments.](/assets/images/guides/nocodb/content-pipeline.png)

<details class="optional-reading" markdown="1">
<summary>Commands behind the deployment</summary>

Create an App, prebuilt-image Environment, and managed data, then link the
logical PostgreSQL database and Redis cache:

    blossom app:create --data app.yml
    blossom environment:create --data environment.yml
    blossom database_instance:create --data postgres.yml --follow
    blossom cache_instance:create --data redis.yml --follow
    blossom environment_link:create --data database-link.yml
    blossom environment_link:create --data redis-link.yml
    blossom env_var:push --environment environment-100 --file .env.production
    blossom service:create --data web-service.yml
    blossom environment:deploy --id environment-100 --follow

Configure `NC_DB_JSON` as a PostgreSQL connection object, rather than a plain
URL string. Map the linked Redis connection to `NC_REDIS_URL`. Preserve the
image's startup script and Node options. Use an HTTP readiness check at
`/api/v1/health` on the web port.

After initialization, save the final `NC_SITE_URL`, create the private worker
with `NC_WORKER_CONTAINER=true`, and deploy both Services:

    blossom service:create --data worker-service.yml
    blossom environment:deploy --id environment-100 --follow

Bucket settings:

```yaml
NC_S3_BUCKET_NAME: your-private-bucket
NC_S3_ENDPOINT: https://your-account.r2.cloudflarestorage.com
NC_S3_REGION: auto
NC_S3_FORCE_PATH_STYLE: "true"
NC_SECURE_ATTACHMENTS: "true"
```

Supply `NC_S3_ACCESS_KEY` and `NC_S3_ACCESS_SECRET` privately. Choose a credential
lifetime suitable for your deployment.

</details>

## Keep Your Records Through Updates

> **Ask your AI agent:** Update NocoDB and check that our records, form, and attachments still work.

> Follow this guide: https://docs.blossom88.com/deploy/nocodb/



Keep the database, bucket, connection settings, and application secrets stable.
Follow NocoDB's supported upgrade path and back up PostgreSQL and your bucket.

The web and worker use external data and storage, so replacing their
Containers doesn't require keeping their local disks. Check upstream worker
and scaling requirements before adding replicas; this guide tests one web
and one worker, rather than establishing horizontal scaling.

## What You Get

- A private spreadsheet-style workspace at an HTTPS address
- Tables, a form, and authenticated API access to your records
- Separate web and worker Services with managed PostgreSQL and Redis
- Attachments in your own private S3 or R2 bucket
