Our team and mission

This position is based at the European Centre for Space Applications and Telecommunications (ECSAT) - Harwell, UK

IMPORTANT: Due to the location of the ECSAT, only the students enrolled in a UK University or students with a UK citizenship or the right to work in the UK can apply to this internship.

Under the direct authority of the Systems, Strategic Programme Lines and Technology Department, the Future Programmes Division is responsible for:

• The management of ARTES Generic Programme Line Future Preparation;
• The scouting, recommendation of innovative and disruptive satellite communication concepts and solutions;
• The elaboration in close coordination with Strategic Programme Lines of new innovative and disruptive satellite connectivity and secure communications concepts and solutions;
• The management of the technology and system exploration (iteration, innovation, disruption) in support of the technology work plans and roadmaps and for the creation of new projects that could lead to creation of new programmes. For example, the implementation of competitions, hackathons, co-creation framework;
• The technical and programmatic consolidation of new telecommunications system projects in partnership with private and/or public entities;
• Leading standardisation, regulatory and spectrum engineering support activities for the Directorate;
• The coordination of functional and integrated support provided by other ESA Directorates and ESA Security Office in collaboration with other departments. 

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

Field(s) of activity for the internship

Topic: Internship in application of NLP technologies for Mission Brainstorming Assistance

Background

The European Space Agency generates a vast amount of information in the form of technical reports, project plans, and other documentation. With the exponential growth of data, there is a need for efficient and effective knowledge management tools that can extract meaningful information from these reports.

Objectives

The objective of this internship is to explore the use of advanced Natural Language Processing (e.g. Large Language Models) for improving knowledge management in satellite communications projects reports. More specifically, the main research objective of this internship would be to develop an AI-based Conversational agent (or ChatBot) focusing on information-retrieval, summarizing and question-answering capabilities.
 

To fine-tune the model with relevant domain-specific knowledge, you will be provided with all the ARTES Future Preparation (FP) reports, which will account for hundreds of documents to be pre-processed and examined by the ChatBot. An important aspect of this research is the potential to expand the scope of the model with other datasets.

.

To that end, you will perform the following tasks

Task 1: State-of-art review
Analyse the strengths and limitations of existing LLMs in the context of satellite communications data reports (ARTES FP reports).

Task 2: Data Pre-processing
Pre-process the data (in the form of pdf and word format) to prepare it for LLM analysis.

Task 3: Build the ChatBot 
Develop an LLM-based conversational agent/ ChatBot that enables question-answering over the dateset in which the LLM has been fined-tuned with

This opportunity will allow you to be exposed to the work done in Future Preparations and Future Projects under the Connectivity and Secure Communications directorate of ESA, as well as to interact with a multi-disciplinary engineering environment covering all Satcom technology domains.

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.

Additional Requirements:

• A foundation in NLP techniques, algorithms and tools (e.g summarization, Q/A, semantic search, keyword extraction, NER, etc.)
• Familiarity with NLP libraries and frameworks (e.g NLTK, spaCy, Transformers, GPT, Bert) is a strong asset
• Experience in programming languages commonly used in NLP (e.g Python)
• Experience in using open-source LLM models (e.g. Llama, Falcon, etc)
• Ability to collect, preprocess and analyze textual data
• Experience with version control systems (e.g., Git) is an asset