Tuning machine learning models, especially finding the right hyperparameters, can be difficult and time-consuming. In addition to the computational effort required, this process also requires some ancillary efforts including engineering tasks (e.g., job scheduling) as well as more mundane tasks (e.g., keeping track of the various parameters and associated results). We present CTE, a general Continuous Training Engine framework to help data scientists speed up model tuning and bookkeeping. With CTE, users can use all available computing resources in parallel for model training. The user-friendly system design simplifies creating, controlling, and tracking of a typical machine learning project. The design also allows researchers to integrate new hyperparameter optimization algorithms. To demonstrate its flexibility, we show how CTE integrates a few major hyperparameter optimization techniques (from random search to neural architecture search).