The Japan/New Zealand/US Microlensing Observations in Astrophysics (MOA) collaboration has been engaged with the detection of exoplanets via gravitational microlensing of over 25 years. This work has resulted in a large data archive of high cadence observations. These field observations need to be reduced, producing light curves of observed objects. The millions of light curves produced will have to be categorized using, for example, machine learning. These categories can be used to (i) classify new light curves and (ii) identify new, unknown classes of objects. The candidate will:Generate light curves for transient objects in the MOA project fields towards the Galactic Bulge and Magellanic clouds using existing photometric software as required, andPerform unsupervised learning to categorize transient light curves,Identify classes of known progenitor objects, such as variable stars, eclipsing binaries, novae etc.,Identify classes of unknown progenitor objects,Investigate one or more classes of light curves.The candidate must have experience in:Python programmingBasic astronomy and astrophysicsExperimental physicsIdeally, the candidate would have experience in:Machine LearningSignal processing, especially time series analysis, fourier analysisData modellingEmail your CV and academic record (including grades and subjects) to the contacts for this position. Also include the GPE estimation calculated used the online calculator linked to in the "Funding Notes" section. Requests for further information or correspondence without all of this information will be treated with low priority and are unlikely to receive a response.