Launched this week
Forex Consensus Engine
The self-hosted "Supreme Court" for your price feeds.
3 followers
The self-hosted "Supreme Court" for your price feeds.
3 followers
The "Multi-Source Paradox" costs banks millions: Bloomberg says 1.05, Reuters says 1.07, and Risk model is flying blind. The Forex Consensus Engine (FCE) is not another data feed. It is a sovereign Arbitration Layer that sits inside VPC. How it works: Ingest: Pipes in existing feeds (Fix/REST/Socket). Police: Uses Context-Aware Physics to detect "Flatlines" and "Flash Crashes" in real-time. Unify: Outputs a single, mathematically defensible "Golden Source" rate for downstream apps.
👋 Makers,
I built the Forex Consensus Engine because I got tired of the "Multi-Source Paradox."
I saw Risk teams using one price (e.g., from Reuters) and Accounting teams using another (e.g., from Bloomberg), simply because their APIs were out of sync by 200ms. This creates a Valuation Gap that fails audits and bleeds P&L.
Most solutions ask you to send your trade data to their cloud API to "fix" it. I believe financial infrastructure should be sovereign.
FCE is different:
On-Premise First: It runs as a Docker container inside your VPC.
Zero Egress: Your trade volume/positions never leave your building.
Physics-Based: It doesn't just "average" the feeds. It uses context-aware outlier detection (Herstatt checks, Flash Crash logic) to reject bad data before it hits your books.
It’s effectively a "Supreme Court" for your data feeds.
Happy to answer questions about the Weighted Confidence Logic, the Docker architecture, or why we chose Python/FastAPI for the core engine!