Python Job: Research Scientist

• APS Levels 6 to EL 1, Ongoing/Non-ongoing
• $82,676 - $112,725 + 15.4% super
• Any Bureau Capital City Office

The Research Program of the Bureau of Meteorology is seeking a scientist with a data / computational science background and physical sciences focus to be a key member of an innovative multi-disciplinary team which improves weather forecast guidance for operational forecast production. The focus of this work is on collaborative development, testing and documentation of weather model Ensemble Post Processing (EPP), including machine learning science, software and systems. The successful applicant will be expected to contribute to the development of new best guess and probabilistic high-impact weather forecast algorithms, science and guidance, utilising many state-of-the-art numerical weather prediction models as inputs. This includes identification and application of predictors and measures of model error to optimally calibrate and add detail to model outputs, including calibration of probabilities of weather events derived from the values and spread of model outputs. The position will also contribute to EPP science testing (verification and evaluation) through the specification and development of verification libraries and collaboration with operational forecast staff. Processing of large datasets will be required. There is a strong focus on developing effective solutions for key operational applications. Work will include collaboration with the UK Met Office on post-processing systems and methodologies. Modern tools and practises such as git, automated testing and deployment and code reviews are required. Professional development opportunities relevant to the work will be available.

The software used for EPP is written predominantly in Python and shell scripts utilising 3rd party scientific and data-handling libraries, and is run on Linux platforms. Much of the verification and display software is written in Python. Development of EPP products will support greater automation of routine weather forecast production and communication of forecast uncertainty to users.

We are seeking an individual with a background in physical sciences and data and / or computational science knowledge. The successful applicant will possess effective team-based programming abilities in relation to scientific problems, including those involving large datasets, data engineering, complex applications and the development of complex algorithms suitable for operational use. Ability to visualise and interpret gridded scientific fields is important. A background in meteorology or related fields is desirable. Experience in applying statistical processes to correct errors would be a strong advantage and machine learning knowledge and experience would be an advantage. A keen interest in applied scientific research and systems development is desirable. This is a great opportunity to work on meaningful scientific software applications for the public good, in a research environment with a commitment to innovation and quality

Position Description - PDF Opens in new window

Position Description - Word Opens in new window