// the find
visualHFT/VisualHFT
VisualHFT is a WPF/C# desktop GUI that shows market microstructure in real time. You can track advanced limit‑order‑book dynamics and execution quality, then use its modular plugins to shape the analysis to your workflow.
VisualHFT is a WPF desktop app for visualizing real-time limit order book dynamics, microstructure analytics (VPIN, LOB imbalance, market resilience), and execution quality. It ships with crypto exchange connectors (Binance, Kraken, Coinbase, etc.) and a plugin architecture for adding custom data sources or studies. The target audience is quants and algo traders who want to observe order flow without building their own visualization layer.
The plugin system is genuinely well-designed — drop a DLL in the plugins folder and the PluginManager picks it up at runtime; the IPlugin interface and base classes are clean enough that the SDK templates actually work. The real-time data bus pattern (HelperOrderBook, HelperTrade as event hubs) keeps the UI and study plugins properly decoupled without a heavy message broker. Performance work is real: custom object pools, BlockingCollection replacing naive Queue, and custom queue implementations that the author claims gave 40% throughput improvement — this is the kind of thing that matters when you're processing 10k order book updates per second. The study library covers the analytics that actually matter for microstructure analysis: VPIN, OTT ratio, and LOB imbalance are not toy metrics.
Still targets .NET 7.0, which is out of support — the .github/upgrades folder full of migration planning docs suggests this is known but unfinished, and shipping an EOL runtime is a real liability for anyone running this in production. The architecture still shows its database roots: there are SQL scripts for table creation sitting in the repo, and the position/analytics layer implies a SQL Server backend that the open-source edition doesn't seem to fully expose or document. WPF-only means Windows-only, which is a hard constraint for most quant teams who run Linux in their colocation infrastructure — you can observe, but you can't deploy this anywhere near your actual trading system. The 'Coming Soon' section has been there long enough that it's hard to trust as a roadmap.