Owen Hobbs
Trading Operations → Algorithmic Systems
Built a complete algorithmic trading system — strategy design, backtesting, optimization, live execution, and AI-powered monitoring — with Claude as development partner.
Results
ProScore2 · Jul 2025 – Jan 2026
Return
—
Win Rate
of trades profitable
—
Sharpe Ratio
risk-adjusted return
—
Max Drawdown
worst peak-to-trough drop
—
How It Works
Strategy Design
Defined trading hypothesis: mean-reversion with funding rate confirmation on BTC-USD perpetuals.
Build Strategy
Implemented matching backtest and live trading code — both generate identical signals from the same parameters.
Parameter Optimization
Systematic parameter sweeps with constraint filtering. 0 combinations tested.
Live Deployment
Deployed to dYdX v4 perpetual futures with automated risk controls and sub-second execution.
AI Monitoring
Active6 Claude-powered agents monitor the system 24/7 — detecting bugs, validating signals, and analyzing performance.
Under the Hood
NautilusTrader
High-performance trading framework (Rust/Python). Event-driven architecture with tick-level backtesting precision.
dYdX v4
Decentralized perpetual futures on Cosmos SDK. On-chain order book with sub-second block finality.
Claude AI
Anthropic's language model powering development partnership, code generation, and the autonomous agent framework.
Python
Core trading logic with async processing, NumPy/Pandas analytics, and type-safe validation via Pydantic.
AI Agent Framework
6 specialized agents monitor the live system 24/7 — bug detection, health checks, signal validation, and more.
What's Next
- -Deeper backtesting rigor — expanding in-sample/out-of-sample validation to stress-test strategies across unseen market conditions.
- -New strategy classes — Order Book Imbalance and orderflow analysis, adding microstructure-based edges alongside mean-reversion.
- -ML integration — machine learning for regime detection and pattern recognition, letting the system adapt to changing market conditions.
- -Autonomous strategy pipeline — I provide the strategy design, AI agents handle the build, backtesting, optimization, and reporting end-to-end.