How the technology works

Cognaize is neuro-symbolic AI built for finance. Neural models read your documents. Symbolic models verify every output against your institution's rules, and the system keeps working until every check passes.

The core idea

Two kinds of intelligence, one system

Cognaize merges two kinds of AI: neural models that adapt to messy documents, and symbolic models that enforce your rules. A workflow engine coordinates them to keep the system efficient at scale.

Neural models LLMs · layout · vision

Models that learn from examples. Language models (LLMs and SLMs) read the text; layout and vision models read tables, structure, and charts.

Symbolic models rules · logic

Hard rules and accounting identities, written as logic. Every check either passes or fails, and the same input always gives the same answer.

Adaptive where documents are messy. Exact where the rules demand it.

The limitation

Why not use an LLM to analyze financial documents?

LLMs reason by similarity, not rules. They place "cash" near "restricted cash" because the words appear in similar contexts. Useful for reading language, unreliable for financial data.

LLMs are not built for finance

LLMs are non-deterministic. Finance requires deterministic outcomes.

Expensive at scale

Running large models on every page of every document gets costly fast at institutional volume.

Your rules

Your ontologies, embedded in the AI

Every institution interprets data differently. Cognaize captures your schema, terminology, and rules, then enforces it on every document.

01

Define your ontology once

Your schema, terminology, and rules, written in plain business terms.

Schema Glossary Rules
02

The Semantic Kernel translates it

Plain language becomes formulas in code the system can run.

03

Every document validated

Applied automatically, on every page. Same input, same output: every check on record.

The pipeline

From document to data

  1. Read

    Finance-tuned models (OCR, layout, vision, SLMs) extract text, structure, and visuals from any document quality.

  2. Structure

    The system creates a version where tables, footnotes, and text become connected elements.

  3. Contextualize

    Field-by-field extraction adjusted to your schema and definitions.

  4. Verify

    Every value is checked against your rules. Any failure triggers a re-run or analyst review.

  5. Orchestrate

    Feed usable data, in any format, to where your systems live (e.g. JSON, Excel, CSV, etc.).

The engine

The cheapest path that passes

The Hypergraph Workflow Engine starts every document on the lightest models and escalates only when a check fails.

Most documents never need the heavier models.

~90% lower per-document cost than LLM-only processing
Built for Enterprise

Intermediate results (OCR text, detected tables) are saved and reused.

The foundation

Models built for finance

Cognaize models are trained on over 8 million fully annotated financial documents, each labeled by CFAs.

8M+ financial documents, fully annotated by CFAs
Small language models Narrative text & classification
OCR engines Scans become text
Layout detectors Tables & structure
Vision-language models Charts & figures
Computer vision Key page regions

See it run on your documents.

Your Rules. Your Schema. Your Data.

Request a demo