Now in early access

Document ingestion
built for teams

Upload, process, and semantically search documents across your entire organisation. Isolated per tenant. Powered by Qdrant vector storage.

100%Tenant isolated
<2sAvg ingest time
RealtimeStatus updates
Documents
Upload
Usage
Settings
Features

Everything your team needs

A complete ingestion pipeline with multi-tenant isolation, usage controls, and real-time status tracking.

🏢

Multi-tenant isolation

Each organisation gets its own Qdrant collection and S3 prefix. Zero risk of cross-tenant data leakage.

Real-time status

Live updates via Supabase Realtime. Watch your documents go from pending → processing → done.

📊

Usage controls

Set per-org limits on document count, storage, and API calls. Enforced at the API layer before upload.

🔍

Semantic search ready

Documents are chunked, embedded, and stored in Qdrant — ready for vector search the moment ingestion completes.

⚙️

Configurable pipeline

Choose chunk size, overlap, and embedding model per organisation. Defaults apply when not configured.

🔒

Row-level security

Supabase RLS ensures every database query is automatically scoped to the authenticated org. No extra filtering needed.

How it works

From upload to searchable in seconds

1

Upload your document

Drag and drop PDFs, TXT, or DOCX files. Limits are checked before anything is stored.

2

File stored in S3

The raw file is saved to your org's private S3 prefix immediately.

3

N8N processes it

Our webhook triggers chunking, embedding generation, and vector storage in Qdrant.

4

Search it instantly

Your document is now semantically searchable. Use your Qdrant collection directly.

Ready to start ingesting?

Get your organisation set up in minutes. No credit card required.

Sign in to your dashboard →