SD-25114 –R&T SCIENTIST IN MACHINE LEARNING

at  Luxembourg Institute of Science and Technology LIST

EDUCATION

• PhD degree in Computer Science, Environmental Science, or similar disciplines
  Experience and skills

Main missions

• The selected candidate will play a central role in the project and its outputs. Her/his main mission is to develop a data-driven crop yield forecasting tool capable of delivering in-season probability information on potential crop yield anomaly and quantity estimation. In a first step, the tool will be tested for targeted geographical areas and specific crop types at province and country level.
• Required Seniority: 2 years of Post-Doc
• Technical Skills: Advanced Statistics and Machine Learning Techniques, EOs, Crop modelling, High Performance Computing

EDUCATION

Doctorate

PROJECT MANAGEMENT TASKS:

• Establish a continuous communication and effective collaboration with the partners of the project.
• Assist in the preparation of project reports and presentations in project meetings.
• Participate actively in the maintenance of a project-dedicated version-control system (e.g., GitLab).
• Explore and employ cutting edge software packages facilitating the interoperability and reusability of the data generated in the project.

DISSEMINATION, VALORISATION AND TRANSFER TASKS:

• Contribute to dissemination, valorisation and transfer of project results (e.g., participation in scientific conferences, exhibition of technology, training sessions, drafting of technical reports, and publication in reputed peer-reviewed scientific journals).
• Participation in the implementation of technological solutions (proof-of-concepts, prototypes).

Scientific work tasks:

• Develop workflow to ingest multiple EO data streams into ML/DL techniques.
• Identify skillful predictors of crop yield forecast at different lead time
• Generate crop yield forecasts at different lead time for the selected case studies of the project.
• Perform robust uncertainty analysis and anomaly outlooks of crop yield forecasts.
• Integrate additional data streams generated by a crop growth model for training ML/DL technique