Open‑source quant finance repos listed

A social post is turning a handful of GitHub projects into a starter kit for quantitative finance, pointing newcomers to code they can fork instead of building from scratch. (github.com, github.com, github.com) Quantitative finance means using code, statistics and market data to test investment ideas before real money is at risk. In practice, that usually starts with three jobs: measuring returns and drawdowns, simulating trades on old data, and checking whether a model still works on new data. (github.com, github.com) One of the repos in the mix, QuantStats, is built for the scorekeeping side of that work. Its documentation says the library calculates metrics including Sharpe ratio, win rate and volatility, plots drawdowns and rolling statistics, and generates HTML tear sheets that summarize a strategy’s results. (github.com) Another, Stefan Jansen’s machine-learning-for-trading repository, is more like a worked example library than a single package. The repository says it contains more than 150 notebooks across 23 chapters, covering data sourcing, feature engineering, long-short strategy design, deep learning and strategy evaluation. (github.com, github.com) Goldman Sachs’ gs-quant sits closer to an institutional toolkit. Its README describes it as a Python toolkit for trading strategies, derivative analysis and risk management, with documentation sections for pricing and risk, portfolios, backtesting, factor models and baskets. (github.com, github.com) The appeal of lists like this is speed. A researcher who wants to test an idea can pull a repo with notebooks, metrics and plotting already wired up, instead of first writing reporting code, portfolio analytics and backtest scaffolding. (github.com, github.com, github.com) The trade-off is that open-source finance code is not a trading edge by itself. Jansen’s strategy-evaluation notes warn that backtesting on historical data can lead to false discoveries if a model is tuned too closely to past conditions, and forward testing is used to check whether in-sample results hold up on new data. (github.com) These repos also sit inside a much larger ecosystem. The Awesome Quant list on GitHub catalogs libraries and resources for quants, showing that the field now has public toolboxes for analytics, data handling, optimization and execution that used to be harder to find in one place. (github.com) For anyone trying to learn the field in 2026, the practical lesson is simple: the barrier to entry is less about finding code than knowing which code solves which problem. The repos highlighted in the post map neatly to three common needs — performance reporting, model examples and firm-style risk tooling. (github.com, github.com, github.com)