\section{Introduction}
\label{sec:introduction}

% TODO content notes:
% - Motivate why backtest correctness matters: a single off-by-one index in a
%   vectorized backtester can silently fabricate alpha.
% - State the concrete finding up front: a 15-hyperparameter "golden kernel"
%   (K12 / Iter12) produced via genetic-algorithm search showed Sharpe 14.49
%   on BTCUSDT 1m data — and the same kernel, run honestly (bar-by-bar, no
%   future information), collapses to Sharpe -0.25.
% - Frame the contribution: not just "we found a bug," but a reusable
%   noise-harness methodology (Section 5) that any quant team can run against
%   their own backtester to detect this class of defect using pure
%   Geometric Brownian Motion data with zero real signal.
% - End with a roadmap of the paper (one sentence per remaining section).