Built for sensitive business data. Not bolted on after.
Archeko handles invoices, contracts, and financial records. Security and data isolation are foundational — not premium features you unlock at a higher tier.
ENTITY ISOLATION
Your entities. Completely separated.
If you run multiple legal entities — a holding company, a construction firm, a property portfolio — each one gets its own isolated data space within Archeko.
Documents, metadata, cost objects, and filing structures are fully separated. There are no cross-entity queries. No shared folders. No accidental data leakage between entities.
- Separate data stores — each entity's documents exist in an isolated partition
- Separate filing structures — folder paths, naming rules, and cost objects are entity-specific
- Separate access — users see only the entities they are authorized to access
- Separate audit trails — processing history is scoped to each entity
This is the same isolation model that banks and regulated industries use — but you do not need an IT department to set it up. It works out of the box.
TRACEABILITY
Every action. Every decision. Logged.
Every document in Archeko carries a complete processing history from the moment it enters the system:
- Uploaded — when, by whom, from what source, with what note
- Read — OCR output generated, text extracted
- Classified — document type assigned, with AI rationale
- Extracted — fields pulled, with confidence scores
- Named — canonical filename generated from template
- Filed — placed in folder path, linked to cost object
- Reviewed — if a human reviewed or corrected any AI decision, logged with the change
This trail is immutable — it cannot be edited or deleted after the fact. When an auditor asks "how did this document get here and who decided it was an invoice?", you have a precise, timestamped answer.
EXPLAINABLE AI
The AI shows its work.
Every classification and extraction decision comes with a visible rationale. This is not a black box — you can inspect the reasoning at every step.
Classification rationale
Why a document was classified as an Invoice vs. a Credit Note vs. a Receipt. The AI cites the content signals it used: header text, amount patterns, vendor identifiers, document structure.
Assignment rationale
Why a document was assigned to Project Alpine vs. Project Bravo. The AI references project codes, vendor relationships, and cost object matching rules from your configuration.
Extraction confidence
What fields were extracted and how confident the AI was in each value. Dates, amounts, and vendor names carry individual confidence scores.
Human Review Protocol
When the AI is not confident, it does not guess. It flags the document for your review and tells you exactly what it was uncertain about. You make the final call — and your correction is logged in the audit trail, improving future accuracy.
DATA PROTECTION
GDPR-compliant. Encrypted. Immutable.
- Encryption at rest and in transit
- AES-256 for all stored data. TLS 1.3 for every connection. Your documents are encrypted before they hit disk and stay encrypted unless actively being processed.
- Immutable document storage
- Once a document is filed, the original file cannot be altered or overwritten. Versioning is append-only. The document you uploaded is the document that is stored — byte for byte.
- GDPR compliance
- Data processing agreements available. Right to deletion honored. Full data portability — export your complete archive in standard formats at any time.
- No training on your data
- Your documents are never used to train AI models. Not ours. Not anyone else's. Your business data stays your business data.
- Data residency
- All data is stored in EU data centers. No data leaves the EU. Processing happens in the same region as storage.
INFRASTRUCTURE
Modern serverless architecture. EU-hosted.
Archeko runs on serverless infrastructure — no single points of failure, automatic scaling, and no maintenance windows that take the system offline.
No downtime for updates
The system is updated without service interruptions. You never see a "scheduled maintenance" window.
Consistent performance
Whether you process 10 documents today or 10,000, the system scales automatically. No queues, no slowdowns during peak usage.
Redundant storage
Your documents are stored redundantly across EU data centers. Hardware failures do not result in data loss.
Nothing to manage
There is no infrastructure on your side. No servers to maintain, no software to update, no IT overhead. You upload documents and ask questions — that is the entire interface.
Ready to see it with your own data?
Start a free trial and upload a real business document. See how Archeko processes, classifies, and files it — with full transparency at every step.
Start free trial