Research review

You have results you need to trust. I reproduce them and test every assumption behind them, from the data and methodology to statistics and implementation. Common problems include look-ahead bias, survivorship bias and understated trading costs.

Written report and code. Fixed scope.

New research

You have a question about markets, strategies or data, such as whether a strategy idea holds up or whether a data set carries information. I define the method up front, carry it out, and report what the evidence does and does not support.

Written report and code. Fixed scope.

Custom work

Anything that isn't a research review or new research, such as point-in-time data pipelines, backtest engines or attribution tools. I also build the tools that put research into practice, such as order generation, execution and reconciliation.

Documented, working code. Project-based or ongoing.

Assumptions are written down before results. Every report states what the analysis cannot show. Uncertain results are labeled uncertain.

Defined projects are quoted at a fixed price after a short scoping call. Ongoing work is billed hourly or monthly. Get in touch →