Hi Product Hunt — I’m Jakub, the founder of QuantJourney. Writing a backtester with AI can look like a weekend project. Building one whose results you can inspect, reproduce and challenge is a very different problem. Too many backtests stop at an attractive equity curve. They do not clearly show how signals became positions, when trades occurred, how transaction costs were applied, how cash and holdings changed, or whether the result survived out-of-sample testing. We built QuantJourney Backtester to make that process more transparent. It is an open-source Python research engine with two complementary paths: • A fast, vectorized engine for portfolio and factor research, ranking models, long/short strategies, allocation, rebalancing and parameter exploration. • An order-aware engine for strategies where orders, fills, execution assumptions, transaction costs, cash and position accounting matter. The project currently includes: • 50 strategy examples • 80+ performance and risk metrics • 25+ charts and diagnostics • Walk-forward and out-of-sample testing • Transaction-cost and execution modelling • Reproducible reports and research outputs • Local execution with your own data • Optional managed market data through the QJ API You do not need an account to install the open-source engine and run it with sample data or your own data. This is an active beta, not a claim that every backtesting problem has been solved. There may still be bugs, incomplete documentation and unsupported edge cases. We would genuinely like people to test it, challenge the assumptions and tell us where it breaks. Our longer-term objective is to support the full path from investment hypothesis, through validation and portfolio construction, to production and live execution. I’d especially value feedback from quants, portfolio managers and Python developers: 1. What would stop you from trusting a backtest produced by this engine? 2. Which missing workflow would make it useful in your actual research process? 3. Where do the documentation or assumptions remain unclear? Thanks for taking a look.

QuantJourney Backtester

Open-source Python engine for defensible portfolio research

Open-source Python engine for defensible portfolio research