Transient classification in the LSST data deluge
LSST will revolutionise time domain astrophysics, detecting over 1 million transient objects per night that will require high efficiency classification to enable astrophysical interpretation. Code and algorithms currently used for time domain analysis are computationally intensive and typically implemented in an interpreted language. There is a need for algorithmic improvements within a modern, cloud based computing environment, e.g. exploiting technologies such as Apache SPARK to provide parallelism, performance and fault-tolerance. The student will be tasked to do this and thereby gain exposure to the most up-to-date techniques as well as participating in LSST science.