# Semantic Object Storage at the Edge

> Store and manage objects at 300+ edge locations with full type safety and zero dependencies. Handle files, JSON, binary data, and media with automatic indexing, AI-native access, and <50ms global latency. Production-grade reliability with automatic backups.

Canonical: https://objects.do/

## In code

```ts
import { Objects } from 'objects.do'

// Type-safe storage with automatic indexing
on.Document.uploaded(async $ => {
  if ($.data.mimeType.startsWith('image/')) {
    const analysis = await $.AI.analyze.Image($.data.url)
    $.Search.index.Document({ id: $.id, tags: analysis.tags })
    const thumbnail = await $.Image.resize($.data.url, { width: 200 })
    $.Storage.put.Object({ key: `thumbnails/${$.id}.jpg`, data: thumbnail })
  }
  $.Slack.send.Message({ text: `Uploaded: ${$.data.filename}` })
})
```

## Frequently asked questions

### What types of objects can I store and how are they managed?

Objects.do supports any file type including images, videos, documents, JSON, binary data, and structured data. Each object gets automatic metadata extraction, MIME type detection, and semantic indexing. Objects are distributed across 300+ edge locations with intelligent caching based on access patterns. Support for versioning, soft deletes, and point-in-time recovery. All objects include automatic compression, encryption at rest, and CDN delivery.

### How do objects compose with other .do services?

All stored objects use semantic types that AI agents and other .do services understand natively. Use $.Subject.predicate.Object patterns to chain operations—for example, upload an image, analyze it with AI, generate thumbnails, and index it for search all in one flow. Objects can trigger workflows, feed into analytics, and be queried by AI agents autonomously. Full type inference ensures compile-time safety across all operations.

### What about performance, bandwidth, and storage costs?

Sub-50ms object access from 300+ edge locations worldwide with automatic routing to nearest location. Intelligent caching reduces bandwidth by up to 90% for frequently accessed objects. Storage costs $0.15/GB/month with automatic compression. Bandwidth is $0.01/GB for cache hits, $0.09/GB for origin transfers. First 10GB storage and 100GB bandwidth are free monthly. Enterprise customers get dedicated capacity and custom pricing.

### Can I handle large files and streaming uploads?

Yes. Support for multi-part uploads up to 5TB per object with automatic chunking and resume capability. Stream large files directly to edge storage without buffering in memory. Progress tracking, pause/resume, and parallel chunk uploads for maximum speed. Automatic retry for failed chunks with exponential backoff. Support for signed URLs with expiration for direct browser uploads.

### How reliable is object storage and what about backups?

Battle-tested across billions of objects stored daily. 99.99% uptime SLA with automatic failover across 300+ global locations. All objects are replicated to multiple regions with automatic healing of corrupted data. Point-in-time recovery for up to 30 days with one-click restore. Automatic daily backups to separate storage tier. Support for immutable objects, legal holds, and compliance retention policies.

## Part of Platform.do

objects.do is one of the primitives of Platform.do. Every .do shares one grammar, Business-as-Code, and one SDK.

- What is Business-as-Code? https://platform.do/business-as-code
- Platform.do: https://platform.do/
- Docs: https://docs.platform.do