pykep is a scientific library providing basic tools for astrodynamics research. Algoritmic efficiency is a main focus of the library, which is written in C++ and exposed to Python using the boost::python library. At the library core is the implementation of an efficient solver for the multiple revolutions Lambert’s problem, objects representing direct (Sims-Flanagan), indirect (Pontryagin) and hybrid methods to represent low-thrust optimization problems , efficient keplerian propagators, Taylor-integrators, a SGP4 propagator, TLE and SATCAT support and more. pykep has been compiled and installed successfully on different platforms. pykep is also present in the Python Index providing some precompiled modules for widely used architectures. pykep has been used by the European Space Agency’s Advanced Concepts Team during different Global Trajectory Optimization Competitions GTOC and several research papers as well as for the optimization of preliminary mission scenarion for the M-ARGO interplanetary cubesat concept.