HN Showcase
Back to browse
GitHub Repository

AI-native framework for building trading systems with polyglot bindings.

214 starsC++

FLOX C++ trading systems framework with MCP

by eeiaao·May 11, 2026·3 points·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidNiche GemBig Brain

Polyglot bindings let you write strategies in Python but execute on a single C++ core.

Strengths
  • •Single C++ core executes strategies written in Python, Node, or Codon without rewrite.
  • •MCP server gives AI agents grounded access to indicators, scaffolders, and docs.
  • •Optional exchange connectors keep the core build lightweight for backtesting-only users.
Weaknesses
  • •Trading frameworks are a crowded, high-stakes category dominated by established players.
  • •AI-native DX is a buzzword-heavy claim that needs real-world agent workflow proofs.
Category
Finance
Target Audience

Quantitative developers and algorithmic traders

Similar To

Lean · Backtrader · QuantConnect

Post Description

FLOX is a C++23 trading framework for building trading systems with polyglot bindings. It provides blocks that may be used for setting up execution pipelines, market data gathering and backtesting. Key idea is to create a production grade framework with great ergonomic and AI-native DX. As a part of FLOX there is an MCP available to make it possible to use iterative loops over strategies development to keep focused without distractions to infrastructure implementation. Curious if anyone used similar approaches and tooling. Open for feedback

Previous Show HN: https://news.ycombinator.com/item?id=44157819

Similar Projects

Infrastructure●●●Banger

Sayiir – A simple durable workflow engine (Rust core, Python/Node.js)

Checkpoint-based recovery beats determinism constraints; embed without separate server, unlike Temporal/Airflow.

Big BrainDark HorseSolve My Problem
ybsoft
103mo ago