OBXID Engine: High-Performance Data Gateway
Autonomous inbound gateway engineered in C++20 for enterprise IT perimeters. Filters, validates and streams industrial EDIFACT data directly into structured, AI-ready JSON Lines (JSONL) format at +40 MB/s end-to-end speed in under 35 MB RAM.
Inbound Perimeter Security & Real-Time AI Readiness
The OBXID Engine functions as an ultra-fast, self-contained inbound security shield directly in front of enterprise ERP and SAP boundaries. It intercepts raw EDIFACT streams, validates segment integrity and converts data on the fly into AI-ready JSON Lines (JSONL).
Stealth Commercial Project — Core source code and proprietary enterprise partner mappings remain confidential.
1. System Architectural Role & Perimeter Protection
The OBXID Engine is designed as a high-speed, self-contained inbound security gateway positioned right before the enterprise IT boundary. It does not replace existing ERP or SAP environments; instead, it intercepts raw EDIFACT streams, catches corrupted segments or syntax mismatches in real time and converts incoming data into clean, structured JSON Lines (JSONL) before downstream systems process it.
2. Core Performance Parameters & Hardware Efficiency
Built in modern C++20, the engine eliminates heavy monolithic converter overhead through maximum hardware efficiency:
- High-Throughput Parsing: Achieves over 35 to 40 MB/s processing speed per CPU core end-to-end from initial file transmission to complete reception.
- Lean Memory Footprint: Bounded to strictly O(1) memory complexity (~24 MB to 35 MB RAM per instance), eliminating the need for expensive server upgrades or inflated cloud instances.
- Air-Gapped Cloud-Native Deployment: 100% self-contained air-gapped architecture provided with a customized Helm Chart for rapid, standardized Kubernetes cluster deployment in minutes.
- Guaranteed Data Sovereignty: Sensitive supplier and enterprise business data never leaves the protected internal infrastructure.
3. No-Code Mapping, Express Debugging & AI Readiness
Transformation rules are managed dynamically through a single central configuration file without requiring recompilation:
- No-Code JSON Rules: Business rules, compliance checks and field assignments are stored as human-readable JSON.
- Express-Debugging Diagnostics: Generates structured, machine-readable error logs on anomaly detection, eliminating manual string searches and saving hundreds of developer hours.
- AI & RAG Pipeline Readiness: Native output in JSON Lines (JSONL) allows direct streaming into vector databases, LLM Retrieval-Augmented Generation (RAG) pipelines and real-time supplier dashboards.
Project Status & Technological Roadmap
Focus on parser core hardening, zero-copy optimization and building an automated test suite for regression-free performance.
Development of the enterprise licensing module and high-availability Air-Gapped Kubernetes cluster orchestration.
Deployment of an intuitive web mapping editor and interactive EDIFACT directory wiki with pre-built downloadable mappings.