Platform

One document engine. Four modules on top.

Slate Cue is a single AI platform that we design, build and run. Documents are read once by a shared engine; each module then does one job with the result.

Shared engine

Everything the modules have in common, built once.

Reading a document well is the hard part. The engine does it the same way for every module, so a value extracted by Parse is the same value Shield checks and Assist quotes.

OCR and layout

Reads text and tables from PDFs, scans and phone photos, and keeps track of where each piece sits on the page.

Indian formats and languages

Understands GST, PAN and IFSC formats, Indian number and date formats, and regional languages.

Field extraction

Turns each document type into a defined set of fields and tables, ready for your systems.

Validation rules

Checks totals, tax arithmetic and identifier formats, and flags values that break the rules.

Confidence and source

Every value carries a confidence score and a link to its page and position in the original.

Human review and audit

Uncertain values go to a person. Every extraction, check and correction is logged.

Architecture

How a document moves through Slate Cue.

Slate Cue architecture Documents arrive through the web app, REST API, WhatsApp or email. Intake checks file type and size, cleans up images, splits pages and detects the document type. The document engine runs OCR and layout analysis, extracts fields, applies validation rules, and attaches a confidence score and source link to every value. The four modules (Parse, Shield, Assist and Recommend) use the engine's output. Results go out through the REST API and webhooks, CSV and Excel, Tally and Zoho exports, and a review screen. Low-confidence fields go to a human review queue. Everything runs inside a security layer: encryption in transit and at rest, role-based access, audit logging and hosting in India. INPUTS INTAKE SHARED ENGINE MODULES OUTPUTS Web app upload REST API WhatsApp Email inbox Intake File type and size checks Image clean-up Page splitting Document type detection Document engine OCR and layout analysis Field extraction Validation rules Confidence scores Source links Audit trail Parse Live / Pilot Shield Early access Assist Early access Recommend Roadmap REST API + webhooks CSV and Excel Tally / Zoho exports Review screen Human review queue Low-confidence fields go to a person. Corrections are logged. Security layer Encryption in transit and at rest · role-based access · audit logging · hosted in India · no training on customer data
Integrations

Fits into the tools you already use.

REST API

Send documents and receive JSON. Webhooks tell your system when a document is ready.

WhatsApp

Customers and staff send documents to a WhatsApp Business number and get replies there.

Tally and Zoho

Export extracted invoices and statements in formats Tally and Zoho Books can import.

CSV and Excel

Download any result as CSV or Excel for your own analysis or older systems.

Request a demo

See Slate Cue on your own documents

Tell us what you work with. We will set up a short demo using sample documents like yours, and talk through a pilot.

  • A short call with the people building the product
  • Free pilot for our first 10 customers
  • Your documents stay private. We never use them to train models.
We use these details only to reply to you. See our privacy policy.