NeuralChainX
Talk to sales
Technology 03 Enterprise RAG

RAG & Security

An enterprise AI security platform that brings Korean-language document handling, vector encryption, and knowledge graphs together

A summary of the core technology for reviewers and practitioners. Detailed specifications are provided in a separate proposal.

01 — Pipeline

The data processing pipeline.

01

Upload

Bulk-load documents with a security level assigned.

02

Document parsing

Parsed by 13 format-specific engines.

03

Chunking

Split into chunks at Korean sentence boundaries.

04

Semantic conversion

Sentences are converted into retrieval vectors.

05

Indexing

Semantic and keyword indexes are built at the same time.

Document parsing engines · 13 formats Tables keep their row and column structure · 97.9% extraction accuracy
PDF HWP HWPX DOC DOCX PPT PPTX XLS XLSX CSV TXT MD Image OCR
02 — Core stack

Retrieval accuracy and information security are designed at the same layer.

A single query passes through four layers in order. Security level determines the retrieval candidates; the quality gate determines the final answer.

L01 Accuracy

Hybrid retrieval

Unified query across semantic search (vectors) and exact keyword matching (BM25)

Multi-document reasoning across a knowledge graph (GraphRAG)

Re-ranking narrows 50 candidates down to 8 precise ones

L02 Confidentiality

RAG-Guard vector security

AES-256-GCM with our own vector transform (SRXT)

Search runs while the data stays encrypted, with no loss of result quality

Per-customer independent encryption keys · processed in a trusted execution environment

L03 Control

Security levels and access control

Four levels — public, department, organisation, confidential — aligned to the NIS N2SF standard

Permissions filtered in real time at query time

Tamper-evident retention of access logs · integrated with a cryptographic verification module

L04 Reliability

Answer quality and safeguards

10-stage quality reinforcement · automatic query expansion (HyDE)

Prompt-injection detection and blocking · automatic de-identification of personal data

Citation verification plus automatic scoring on three answer-quality metrics

03 — Deployment

Your security policy decides how it runs

On-premises

Documents and models both run on your own servers. Air-gapped deployments are supported.

Maximum control
Hybrid

Documents stay in-house and only inference is separated. Routes are split by security level.

Balanced
Dedicated SaaS

Start immediately on a tenant dedicated to your company. Keys and indexes are kept separate.

Fastest to start
Model choice

Choose between our own lightweight model (xLLM) and open LLMs, and assign a different model to each security level.

Throughput

Large document sets are processed concurrently, and repeat queries are answered straight from cache.