Postdoctoral researcher in Science Operations

Skatelescope; Manchester; Remote; GBP 35,000 - 45,000; Full time

Benefits offered by this job

Conference travel

Flexible working hours

Job summary

SKAO Science Operations invites an early-career researcher for a 1-year fixed-term Postdoctoral position to explore and prototype ML/AI techniques for radio observatory operations. The role is based at the SKAO Global Headquarters, Jodrell Bank, UK, with collaboration across scientists, engineers and software developers.

The successful candidate will be able to work independently and in a globally distributed team, with potential to travel for conferences.

Qualifications

• University PhD or near submission in Astronomy, Physics, CS, Data Science, or related field.
• Expertise in radio astronomy or ML/AI in physical sciences.
• Ability to communicate technical work clearly to diverse audiences.
• Experience working independently and in globally distributed teams.

Responsibilities

• Identify and prioritise ML/AI opportunities to improve observatory operations.
• Develop, train, test and validate ML/AI models using real and simulated data.
• Evaluate performance, robustness, explainability, and reproducibility.
• Collaborate with software/engineering teams to integrate ML/AI tools.
• Document methods, results and recommendations.
• Present results at internal meetings and to the wider community.
• Undertake personal research up to 25% of employed hours.

Skills

ML/AI expertise

Radio astronomy knowledge

Communication skills

Team collaboration

Time-series analysis

Anomaly detection

RFI detection

Education

PhD in Astronomy or Physics or CS/Data Science

Job description

An exciting 1-year opportunity for an early-career researcher with expertise in machine learning and artificial intelligence to help shape how these techniques are used in the operations of a world-leading radio observatory.

The SKAO Science Operations group is seeking to employ a talented and motivated Postdoctoral Researcher to join our team based in the SKAO Global Headquarters located at Jodrell Bank in the UK. This is a fixed-term position for 1 year.

With both SKA-Low and SKA-Mid now in the construction and commissioning phases, planning for operations has entered an exciting phase. Modern radio telescopes generate data volumes and operational complexity on a scale that makes machine learning and artificial intelligence increasingly attractive tools. This role will explore, prototype, and evaluate ML/AI techniques to support efficient, reliable observatory operations in areas such as observation scheduling, telescope and data quality monitoring, anomaly and RFI detection, and support for our user community.

The successful candidate will be highly skilled in communication, as well as in collaborative and inclusive work practices. They will work closely with scientists, engineers and software developers to turn operational needs into well-defined, testable ML/AI solutions. To facilitate this we are seeking a researcher with either a PhD in radio astronomy and practical experience with ML/AI tools, or a PhD in computer science or equivalent fields with demonstrated expertise in implementing ML/AI techniques for the physical sciences. PhD candidates nearing submission will also be considered.

The appointee will be encouraged to carry out research up to 25% of employed hours. Appropriate support, including conference travel, will be provided.

This is a one-year fixed term position and is based at the SKAO Headquarters at Jodrell Bank, Cheshire, UK.

SKAO is committed to providing an inclusive and flexible working environment, meeting the requests of our Colleagues whilst also fulfilling the needs and objectives of the Observatory. This role may be offered on a full-time or part-time basis. Please raise your preference in your application.

Due to the locations involved in the SKA project, in particular the UK and Australia, this role requires the post holder to work across different time zones and, in line with SKAO policy, flexible working hours will be supported in agreement with the line manager.

Informal enquiries may be made to Sarrvesh Sridhar ([e-mail]).

The successful candidate will support the future science operations of the SKAO in the following ways:

• Identify and prioritise opportunities where ML/AI techniques can improve the efficiency, reliability and quality of SKAO observatory operations;

• Develop, train, test and validate ML/AI models and prototypes using commissioning, simulated and archival radio astronomy data;

• Evaluate the performance, robustness and limitations of ML/AI approaches against operational requirements, including explainability and reproducibility;

• Work with software and engineering teams to understand how ML/AI tools could be integrated into SKAO operational software and workflows;

• Contribute to the development and refinement of the Operations workflows where it relates to the use of ML/AI;

• Document methods, results and recommendations clearly so that the work can be continued and built upon;

• Effectively collaborate with SKA staff across the three sites;

• Communicate openly, effectively and respectfully with internal and external stakeholders, and present results at internal meetings and to the wider community;

• Optionally undertake a programme of personal research up to 25% of employed hours;

• Undertake any other duties as directed.

The post holder will have:

• A University PhD degree (or close to submission) in Astronomy, Physics, Computer Science, Data Science or another closely related and relevant field, or equivalent qualification.

• Either (a) expertise in radio astronomy techniques together with practical experience of applying ML/AI tools, or (b) demonstrated expertise in implementing ML/AI techniques to solve problems in the physical sciences.

• Demonstrated ability to communicate technical work clearly, in writing and verbally, to audiences that include scientists, engineers and non-specialists.

• Ability to work both independently and collaboratively within a globally distributed team.

• Experience applying ML/AI to time-series, imaging, anomaly detection, classification or optimisation problems.

• Familiarity with radio interferometric data and processing (e.g. visibilities, calibration, RFI flagging), or a willingness to learn quickly.

• Experience with software engineering best practice, including version control, testing and deploying models in production or near-production environments.

• An appreciation and understanding of the science cases motivating the SKA Observatory.