Our team and mission

This position is based at the European Space Research and Technology Centre (ESTEC) – Noordwijk, Netherlands

The Software Systems division covers the domain of software engineering for embedded software systems (i.e. software on-board a spacecraft) and ground software systems (i.e. software systems to support the design and verification of space systems). The division is concerned with the necessary Software engineering technologies, methods and tools to support the development of such software systems. This covers in particular standards and reference architectures, System-software co-engineering methods, Requirements engineering and modelling, Design methods, automatic code and test generation, languages and compilers, Modelling and Simulation for system engineering and in support of testing and verification.

In addition, the division studies the challenges of new technologies (such as artificial intelligence, machine learning, Big Data, …) and the possibilities of their applications for space systems.

For further information visit our web site: http://www.esa.int

Field(s) of activity for the internship

You can choose between the following topics:

1) Topic 1: Uncertainty quantification for AI space applications

The objective of this opportunity is to identify the most promising techniques and open-source frameworks related to robustness and uncertainty quantification methods, identify a space related application, implement the identified techniques to improve robustness and uncertainty quantification, evaluate the possibility to use these methods on-board.

.

2) Topic 2: LLMs applied to static Analysis

The objective of this opportunity is to identify the limitations of static analysis and to evaluate how LLMs can be used to complement static analysis in the code review process. For this task the recently released Code Llama, available through a permissive license, is initially identified as the LLM to be used. The intern will also identify the best way to deploy the model for an optimal use, considering a local deployment through a quantization version.

.

3) Topic 3: Generating System Requirements Document with Natural Language Processing

The System Requirements Document (SRD) is a key document defining a mission’s system function, overall system performance, system segment objectives, and interfaces. Each mission has its own SRD, but all SRDs share similar features and sometimes even content. To accelerate the definition of requirements in the early design phases, ESA is developing a generic SRD, with a manual, top-down approach.

The goal of this internship will therefore be to investigate Natural Language Processing (NLP) methods (including Large Language Models) to semi-automatically generate a set of generic systems requirements for space missions, thus complementing the top-down approach.

In parallel, ESA, through the OSMoSE initiative, has been leading the effort to define an ontology of space systems, the SSO (Space System Ontology), facilitating the data exchange between the various stakeholders of the space field. The generated set of requirements will have to be aligned with the concepts from the SSO. Through this activity, the intern will have the opportunity to apply NLP/LLMs to space systems requirements generation while discovering the SSO and exploring various conceptualisation approaches.

Behavioural competencies

Result Orientation
Operational Efficiency
Fostering Cooperation
Relationship Management
Continuous Improvement
Forward Thinking

Education

You must have student status and be enrolled at university for the entire duration of the internship. You should preferably be in your final or second to last year of a university course at master’s level in a technical or scientific discipline.

Additional requirements

The working languages of the Agency are English and French. A good knowledge of one of these is required. Knowledge of another Member State language would be an asset.