Artificial Intelligence Software Engineer

The ESRF hosts many beamlines which allow the data collection of large datasets (up to millions of one or two-dimensional datasets per day), which require modern strategies for fast (high-throughput) and big data analysis. The objective of these developments is firstly to enable smart procedures to extract relevant data from datasets too large to be evaluated individually, and secondly to enable automated data acquisition by detecting on-the-fly which areas of a sample should be studied at a higher resolution.

To achieve these goals you will implement state-of-the-art algorithms based on artificial intelligence and deep learning. One part will consist of developing algorithms for offline data analysis & classification, which can be tested on a variety of techniques (X-ray fluorescence or powder diffraction) with similar data structures. The second part will consist in integrating these algorithms in the data acquisition procedures (all written in python).

The development will take place among a team of scientific software engineers and scientists, and will exploit the ESRF state-of-the-art GPU compute clusters. Further information about the post can be obtained from Vincent Favre-Nicolin (favre@esrf.fr) and Vicente Sole (sole@esrf.fr).

Your missions include:

• the development, implementation, and optimization of data processing modules based on artificial intelligence and deep learning techniques
• the definition of APIs to ensure the inter-operability of the different software components
• the integration of the components to enhance

Documenting your developments to ensure future maintainability is essential.