Strategy research
Systematic strategy development across asset classes: derivatives pricing, second-order risk, volatility structure, and the statistical work that separates a real edge from a fitted curve.
Systems
Research that survives contact with a real market.
The distance between a result in a notebook and a system running against a live venue is most of the work. This is how that distance gets closed.
01 / Capabilities
What gets built
Four areas of work, usually engaged together rather than alone.
Backtesting & data
Event-driven engines and market data pipelines built so a backtest resembles what actually happens at the exchange: real fills, real costs, correct point-in-time data, no lookahead.
Live execution
Order routing, position and risk management, reconciliation and monitoring. Broker and venue integrations designed around disconnects, partial fills and the sessions nobody planned for.
Rebuilds & audits
Systems that half work, and nobody is certain why. Establish what is actually happening, document it, and return the system to a state a team can operate and maintain.
02 / Standards
How the work is done
Method matters more than any single strategy. These are non-negotiable.
- Out of sample, or not at all
- Chronological splits, walk-forward evaluation, block bootstrap for inference. Time series are never randomly shuffled, and a result that only exists in-sample is not a result.
- Costs before conclusions
- Spread, slippage, commission, financing, assignment and borrow are priced in before anything is called an edge. Gross performance is treated as an intermediate number, never a finding.
- Production is the test
- Deliverables run against a live venue with structured logging, reconciliation and explicit failure paths. A strategy that cannot be operated has not been finished.
- Evidence over narrative
- Descriptive work comes before models, and hypotheses are stated before they are tested. Where the evidence does not support a strategy, that is reported as the result.
03 / Infrastructure
Where it runs
Platforms and data sources the systems are built against.
Languages & core
PythonC#Pine Script
NumPy / pandasArrow / ParquetPostgreSQL
Brokers & platforms
Interactive BrokersQuantConnect / LEAN
NinjaTrader 8TradovateTradingView
Market data
DatabentoCME / CBOE
Options chains & surfacesTick & minute history
Platforms and data sources the systems are built against. All names and marks belong to their respective owners; their appearance here indicates tooling in use and does not indicate partnership or endorsement.