Impact
What changed
Turns a trading idea into a controlled data and validation workflow with quote-quality checks, sensitivity testing and reproducible backtest outputs.
Personal project
A strategy validation environment for testing pre-earnings options ideas before considering them for live decision support.
Impact
Turns a trading idea into a controlled data and validation workflow with quote-quality checks, sensitivity testing and reproducible backtest outputs.
Role
Defined product scope and validation discipline; built the workflow with AI-agent collaboration and local deterministic evaluation code.
Status
This is less a claim about being an options expert and more evidence of structured validation: hypotheses first, deterministic evidence second.
Essence
The project focuses on proving or rejecting pre-earnings volatility strategies before they are used for recommendations.
Stage-2 data work moved raw ThetaData artifacts into a DuckDB foundation with event, quote and reconciliation tables.
The Strategy Lab exposes timing windows, side modes, quote offset controls, spread filters and minimum-feasible-trades gates.
Outputs include comparison exports, run summaries, sensitivity checks and Monte Carlo-style validation where sample size allows it.
Technology and concepts