OjasBase.
Start deploying
The OjasBase platform

An operating system around your application.

Connect source, runtime, data, networking and recovery in one application graph. Keep the project—not the server—at the center.

Connected resources

One application. A connected operational picture.

Application graphILLUSTRATIVE / PILOT
GitHub / ZIP immutable source
OjasBaseUnderstand · Build · Operate
Domain + HTTPSActive
Source connected
Selected resource

API

Application runtime with health checks and release history.

State
Healthy
Region
Bengaluru 1 · Pilot
Visibility
Public service
Reserved resources
0.5 vCPU · 1 GB
Compute / storage estimate
₹1,460 / month
Illustrative application. Example prices assume 730 runtime hours, exclude GST and plan minimum adjustments.
Console-quality flow

Useful defaults. Clear progress.

Try project naming and an example deployment walkthrough. Real source analysis and deployment run in the customer console.

Review the detected plan.

Interactive example
A name your team recognizes, such as “Client Portal”.Lowercase letters, numbers and hyphens. Example: client-portal.
Build
npm run build
Start
npm start -- -p ${PORT}
Port
3000
Deployment walkthrough

Uploading source

Advance through an example release. No source is uploaded and no application is deployed.

  1. 1Uploading sourceCurrent stage
  2. 2Storing snapshot
  3. 3Building
  4. 4Starting
  5. 5Health checking
  6. 6Live
Build / Run / Store / Connect

The layers work together.

01

Immutable releases

GitHub or ZIP source becomes a snapshot. Deployment history preserves the release identity and build outcome.

02Pilot

Managed data

Attach Postgres, Redis and S3-compatible Ojas Store resources. Stateful services remain pinned to their node.

03Controlled pilot

Domains + HTTPS

Generated endpoints and verified custom domains connect to healthy releases. Custom domains require enabled support.

04Pilot

Recovery

Keep the previous healthy release, retry failed jobs and roll back. Database backup and restore follow managed resource workflows.

Ojas Postgres

Your application uses the database. We manage the lifecycle around it.

Persistent data has its own lifecycle. A new application release should reuse the database; database deletion remains a separate protected operation.

Pilot

Create. Import. Recover.

Create a private managed Postgres resource. Import an existing database using a PostgreSQL custom-format dump, then verify your schema and data. Keep backups and restore workflows separate from release rollback.

Implemented source · Pilot

Attach. Browse. Query.

Bind a named database to the selected environment as DATABASE_URL. Browse schema and rows, run bounded read-only SQL, and use authorized write mode or primary-key-based row editing. Live-host release gates still apply.

Ojas Postgresproduction-db
Illustrative product view · Pilot
PRODUCTIONPrivateHealthyBengaluru 1 · Pilot
DATABASE_URLproduction-db
SQL EditorRead-only
SELECT id, email, created_at
FROM users
ORDER BY created_at DESC
LIMIT 100;
Example result · 100 rows · 42 msIllustrative timing. No query runs on this website.

Tables + schema · Bounded queries · PK-based row editing

Read the database guide
Dump / SQL import · Backups + restore · Authorized connection
Illustrative binding model · Pilot
Production APIDATABASE_URLproduction-db
Staging APIDATABASE_URLstaging-db

Choose each attachment explicitly. Staging and Preview must never inherit a Production database silently.

Application migrations update the schema. Manual SQL and imports are different workflows. Database Studio is implemented in source for controlled Pilot rollout, subject to host validation.

Connected managed data

Postgres, Cache and Store belong to the application.

01Pilot

Ojas Postgres

Create, attach, browse, query, import, back up and restore persistent application data.

02Pilot

Ojas Cache

Attach a private application cache through an explicit REDIS_URL environment binding.

03Pilot

Ojas Store

S3-compatible object storage abstraction with project-scoped credentials and signed URLs.

Placement / capacity

Reservations keep workloads within safe capacity.

Worker facts, system reserves and active reservations determine availability. Capacity pressure blocks new placement while keeping current reservations.

For application owners

Request CPU and RAM, choose a region and follow placement status. Health, cost and service connections remain visible.

For platform operators

Review capacity demand and provider budgets. Provisioning and maintenance require explicit operator actions. Autonomous scaling and automatic stateful migration are upcoming.

Operational Intelligence

Recommendations backed by runtime evidence.

OjasBase records deployment outcomes and resource samples. Recommendations carry evidence and confidence; changes are review-gated.

ILLUSTRATIVE RUNTIME RECOMMENDATION Approval required
Observed
Memory +41% after the last release
Suggested action
Increase service memory to 768 MB
Confidence
Medium · 14 runtime samples
Traffic controls

Practical controls. Controlled rollout.

01Opt-in pilot

Request limits

Opt-in shared rate limits and bounded request/message size.

02Opt-in pilot

Basic WAF

A URI-pattern filter with monitor/block modes. It is not an enterprise WAF.

03Opt-in pilot

Static cache policies

Conservative cache headers for fingerprinted public static assets. No shared global cache is claimed.

Global CDN, global DDoS protection and autonomous multi-region operation are not offered.

Your next application

Build the product.
Keep infrastructure from
becoming the product.

Start with one application. Add data, domains and capacity as it grows.

Start deployingRead the docs