Back to projects
High-performance commodity trading simulation
GPU-Accelerated Backtesting Engine
Massive speedup vs CPU baselines
Built a CUDA-powered backtesting engine for commodity trading strategies, enabling large-scale parameter sweeps and fast financial metric computation.
C++CUDAParallel Computing
Build Highlights
Parallelized thousands of strategy configurations using GPU kernels.
Optimized memory access patterns and implemented parallel reductions.
Computed key financial metrics including PnL, Sharpe ratio, and max drawdown efficiently.