C++
Core language for latency-critical components
We build systematic trading strategies through rigorous engineering thinking, viewing markets as complex systems, and empowering strong autonomous teams
We define problems clearly, write code to explore them, run simulations, and look at results. Every part of the process is measurable - from model behavior to system latency to logs in production.
Markets operate under constraints, dynamics, and feedback. We build models and infrastructure that account for latency, liquidity, volume shifts, and cost. These aren’t external factors — they’re part of the system we’re solving.
We take full responsibility for what we build. From idea to production, we test, deploy, monitor, and improve things ourselves. We enjoy working this way - it’s clear, honest, and rewarding.
State-of-the-art but pragmatic. We prioritize low dependencies and in-house systems built for reliability and performance.
Core language for latency-critical components
Research, backtesting, and orchestration
Structured data storage and analysis
Cloud infrastructure and compute resources
Containerization and deployment
Real-time monitoring and visualization
We break down market behavior into measurable problems, build systems to test ideas, run simulations, and evaluate results under real constraints. Every stage is observable — from execution quality to latency, market impact, and production logs.
Machine learning in trading isn’t about applying the latest models. It’s about understanding what data can reveal, designing reliable pipelines, and building systems that remain stable under changing market conditions.
Infrastructure isn’t just about speed. It’s about reliability under pressure, observability when systems fail, and architectures that remain maintainable as complexity, scale, and throughput increase.
Quantitative research is driven by evidence, not intuition. We analyze market structure, test assumptions rigorously, and make decisions based on measurable statistical behavior and repeatable results.