17 Python libraries for quants

Quant Science posted a compact, curated inventory of 17 Python libraries aimed at people building hedge‑fund style, production‑capable quant stacks. (threadreaderapp.com) The thread calls out tools across four practical layers: numerical engines and model libraries, options‑and‑derivatives toolkits, optimization and portfolio libraries, and higher‑level backtesting and production frameworks. (wilsonfreitas.github.io) Some named projects are familiar: QuantPy, a small framework with portfolio and Sharpe‑optimization utilities; FinancePy, a derivatives‑pricing library that uses Numba for speed; tf‑quant‑finance, Google’s TensorFlow‑based library of differentiable pricing and simulation routines; and optlib, a Python options‑pricing toolkit with Black‑Scholes, Greeks and chain fetching. (quantpy.readthedocs.io) (pypi.org) (github.com 1) (github.com 2) Those libraries are the sort of building blocks quants stitch together. QuantLib and its Python bindings or GS Quant can supply robust pricing and risk primitives; a differentiable engine like tf‑quant‑finance lets you compute gradients through Monte‑Carlo or PDE solves, which matters when you use gradient‑based optimizers or train neural nets on pricing targets. (quantlib.org) (github.com) GS Quant is a clear example of the trend Quant Science flagged: a major sell‑side desk has open‑sourced the toolkit it uses for derivatives pricing, analytics and Marquee API access, so teams outside the bank can reuse real production code rather than starting from scratch. (developer.gs.com) For a practitioner moving from finance into quant trading, the list is useful in two concrete ways. First, it short‑circuits the scouting process: instead of hunting GitHub for a volatility surface class or an implied‑volatility solver, you can start from known repos and their docs. (wilsonfreitas.github.io) Second, reproducing a small pipeline end‑to‑end—data ingestion + a pricing primitive from FinancePy or QuantLib + a backtest harness—becomes a demonstrable portfolio piece for interviews. (github.com 1) (github.com 2) How these pieces work together in a minimal quant stack is straightforward to picture. A market‑data pull fills a pandas table, a pricing library turns that data into model prices and Greeks, an optimizer or optlib routine builds allocations, and a backtester validates the signal before you wire it to execution software. (pypi.org) (github.com) The immediate payoff of the thread is practical: it’s a checklist for assembling a research‑to‑production pipeline and a cheat‑sheet for the libraries you should master. The thread was posted by Quant Science on March 19, 2026 and links to the 17 repositories and examples so you can clone, run the notebooks and start combining modules. (threadreaderapp.com)