Bolt.new
Quick to scaffold an upload, folder and share-link web app.
Storage · updated 2026-11-03
Don't: No, use a product No for your real files. Sync clients and data durability are where a homemade version loses data; build a simple upload-and-share app only to learn.
An upload page with folders and share links on top of object storage is easy and genuinely useful. Dropbox is something else: desktop clients that sync changes in both directions across devices, resolve conflicts, survive crashes mid-transfer and never silently delete your work. Getting that wrong loses data, the one thing a storage product must not do. Keep your important files on a maintained service and back them up separately.
Dropbox is cloud file storage with desktop sync, sharing links and version history. This page is about building your own version for personal or internal use, or to learn how this kind of product works, not about copying Dropbox to compete with it. How we grade.
Difficulty is for a solo builder using an AI coding tool. Skip means not worth building yourself.
| Feature | Difficulty | Why |
|---|---|---|
| Web upload and download | Easy | Presigned upload URLs straight to S3 or R2. |
| Folders and previews | Easy | A files table with parent_id; previews for images and PDFs. |
| Share links | Easy | Random tokens with expiry, optional password and download limits. |
| Resumable large uploads | Medium | Multipart uploads with retry per part. |
| Version history and trash | Medium | Keep prior versions and soft deletes with a retention window. |
| Desktop two-way sync | Hard | File watching, change detection, conflict copies and crash recovery; the core of Dropbox. |
| End-to-end encryption | Hard | Easy to get subtly wrong; use audited libraries if you try. |
| Mobile apps with camera upload | Skip | Use the platform's photo backup instead. |
Your own version vs Dropbox, at personal or small-team scale.
Plus: $9.99/mo billed annually
Plus for one person with 2 TB, widely listed at $9.99 per month billed yearly or $11.99 monthly; the official page showed EUR 9.99 per month from our location.
Source: dropbox.com/plans. Prices change; check before deciding.
| Running cost | Per month | Notes |
|---|---|---|
| Object storage (Cloudflare R2) | $1–$30 | About $0.015 per GB per month with no egress fees; 2 TB would cost around $30. |
| Hosting and database | $0–$25 | Free tiers for a personal app. |
| Separate backup | $2–$10 | A second copy elsewhere, because one bucket is not a backup. |
| Total | $3–$65/mo | Saves up to $84 a year vs Dropbox Plus; at the top of the range it costs $660 a year more. |
Quick to scaffold an upload, folder and share-link web app.
Good for presigned multipart uploads and careful permission checks.
Fast UI on Supabase Storage for a learning version.
Paste it into your tool, then ask for a plan before any code. Edit the specifics to match what you need.
This is a learning project, not where my important files will live. Build a simple personal file locker with upload, folders and share links. Stack: Next.js, Supabase auth and Postgres, Cloudflare R2 (S3 compatible) for file storage. Data model: folders (id, owner_id, parent_id, name), files (id, owner_id, folder_id, name, size, mime, r2_key, sha256, version, created_at, deleted_at), versions (file_id, version, r2_key, size, created_at), shares (id, file_or_folder_id, token, expires_at, password_hash, max_downloads, downloads). Requirements: - Upload with presigned multipart URLs generated by the server; resume failed parts. - Verify the SHA-256 checksum after upload before marking the file complete. - Folder tree, image and PDF previews, rename, move. - Keep old versions on overwrite; soft delete to a 30-day trash. - Share links with expiry, optional password and download limit. Security: - Bucket must be private; downloads only through short-lived signed URLs after a permission check. - Row Level Security on folders, files and shares. - Share tokens are 32 random bytes; passwords hashed with a slow hash. - Serve user files with Content-Disposition attachment from a separate domain so uploaded HTML cannot run as my app. - Enforce size limits and per-user quotas.
AI tools often get these wrong by default. Check each one before anyone else uses the app.
They went far past the prototype: support, payments, reliability and years of iteration. Their profiles show what that took.
Colin Percival built Tarsnap
Encrypted, deduplicated online backup run by its creator for years.
Aleksandar Jovanovic built SecretDrop
Client-side encrypted, auto-expiring file sharing for secrets.
If you need Dropbox's job done rather than a project, these are built and supported by independent makers.
tarsnap.com
Encrypted, deduplicated online backup service for UNIX-like operating systems, stored on Amazon S3.
secretdrop.dev
Client-side encrypted, auto-expiring file sharing for API keys and .env files.
You learn presigned uploads, object storage, checksums and the permission model behind share links.
Don't. No for your real files. Sync clients and data durability are where a homemade version loses data; build a simple upload-and-share app only to learn.
A working personal or internal MVP takes about 4–12 h with a tool like Bolt.new. A polished version you would rely on every day takes about 38–125 days, mostly spent on two-way desktop sync with conflict resolution across devices.
Expect $3–$65/mo at personal or small-team scale (object storage (cloudflare r2), hosting and database, separate backup). Dropbox Plus costs $9.99/mo (not independently verified), so the most you would save is $84 a year, before counting your time.
You can build a web app for uploading, organising and sharing files on object storage in a day. Desktop sync that never loses data is the hard part, so do not trust a homemade version with your only copy.
For raw storage, often yes at small volumes, since object storage is priced per GB. You give up sync clients, version history you did not build and support, and you must back up the bucket yourself.
Mature open-source options such as Nextcloud or Syncthing handle sync and versioning and have years of fixes behind them. They are safer than a fresh build for real files.
Building your own tool that solves the same problem is common and generally fine. Copying Dropbox's code, design assets, brand name or content is not. Use your own name and design, respect their terms of service, and get proper advice if you plan to sell it. This page is not legal advice.