Intern in the AI and Data Science section

Our team and mission

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

The Future Engineering Division (in support of the program directorates) is responsible for developing and providing the right Engineering Tools, Methods and Standards for the development of future projects. As part of this division, the AI and Data Science section is leading the Agency centre of excellence for Artificial Intelligence and data science technologies for space applications and downstream applications. It supports the identification, selection, adaptation, evaluation of AI technologies and techniques for their use in the development and exploitation of future space missions.

Candidates interested are encouraged to visit the ESA website: http://www.esa.int

Field(s) of activity for the internship

You can choose between the following topics:

1) Topic 1: Building a Knowledge Graph of ESA’s AI activities
Join us on a mission to build a Knowledge Graph (KG) of ESA's AI activities! The KG, or graph database, will store metadata for all AI activities led by ESA. With the holistic view provided by the KG, we will be able to extract trends and assess technological gaps. During this internship, you will first define the ontology structuring the KG and populate the graph database. Then you will explore methods like Large Language Models (LLMs), or other techniques, to predict new relationships within the graph, and infer trends. The KG may also be integrated into a Graph Retrieval-Augmented Generation (GraphRAG) pipeline. You will also implement an intuitive interface to enable users to query the graph in natural language. This project offers an exciting opportunity to apply cutting-edge AI technologies to ESA’s internal data.

2) Topic 2: Automated Planning and Scheduling for a Crewed Lunar Outpost
Automated Planning and Scheduling is the automated process of determining a sequence of actions and allocation of resources to achieve specific goals efficiently. Your task is to implement an automatic scheduling system to organise the daily and weekly tasks of a human and robotic crew living on a moon base. You will first have to identify the tasks (i.e., running experiments, doing housekeeping, or exercising) to be performed by the crew. You will evaluate the current state-of-the-art automated planning and scheduling methods. After selecting your preferred approach, you will design and test the scheduler. Naturally, the crew cannot blindly rely on your scheduler as their survival is at stake! Therefore, the scheduler must be capable of justifying its decisions using Explainable AI techniques.

3) Topic 3: Unlock the Future of Autonomous Space Systems with LLM Agents
Join us at the European Space Agency to explore how cutting-edge Large Language Model (LLM) agents can transform the space domain. The future of autonomous systems is evolving rapidly, with intelligent agents able to solve complex tasks by iteratively working step-by-step toward optimal solutions.
During this internship, you will dive into the current state-of-the-art LLM agent technologies, identifying the most promising frameworks with the potential to be implemented in space-based, embedded systems. Together, we'll explore key use cases where agentic workflows can have the greatest impact, and you'll develop a proof-of-concept (PoC) to showcase the potential of these intelligent agents in space.
If you're passionate about AI, space exploration, and innovation, this is your opportunity to help shape the next frontier of autonomous systems!

4) Topic 4: Implementation of a Semantic Segmentation demo in a Mars Environment

ESA plans to go to Mars, and to do so we want to be able to test our early ideas in a realistically simulated Mars environment. This internship is composed of two parts: designing a playable 3D Mars Environment in Godot from the point of view of a rover, and implementing a semantic segmentation algorithm to parse accurate information from this environment. This work will be used to generate datasets, to inform navigation software, amongst other practical uses. For those who enjoy video game programming as well as machine vision, this internship is for you! You will have access to ESA resources and support during this internship to bring your vision to life.

5) Topic 5: Building a playable Mars Environment
ESA plans to go to Mars, and to do so we want to be able to test our early ideas in a realistically simulated Mars environment. If you are game designer with a vision, this is your chance to implement it! During this internship, you will model an ESA asset base in Blender, to then design a game environment in the open-source game engine Godot. The environment will be navigable by a playable rover and astronaut, ideally accounting for Martian physics and that allows for automated rover navigation.
You will also have at your disposition existing ESA 3D asset libraries in different formats as a starting point. This environment will be used in the future for demo-ing computer vision algorithms and testing different ESA software.

6) Topic 6: AI at the Edge in Space
In the past five years, the number of hardware and software stacks capable of deploying AI at the edge has skyrocketed. While the European Space Agency (ESA) has evaluated various solutions, a clear overview of the most promising commercial off-the-shelf (COTS) and non-COTS options for space is still missing.
As an intern, you’ll dive into this exciting landscape by analysing different COTS and non-COTS solutions for applying AI at the edge in the space domain. You’ll classify these technologies by mission class and criticality. Your work will include a state-of-the-art (SOTA) review of both published research and internal ESA findings, as well as benchmarking AI applications on space-relevant hardware and software architectures.
The outcome? Together, we’ll create a comprehensive survey paper to identify key AI solutions for edge computing in space; an important resource for the space tech community.

Behavioural competencies

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

For more information, please refer to ESA Core Behavioural Competencies guidebook

Education

You must be a university student, preferably in your final or second-to-last year of a university course at Master’s level and you need to remain enrolled at your University for the entire duration of the internship.

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:

Topic 1: Building a Knowledge Graph of ESA’s AI activities
Experience in AI applications development and Python is needed. Knowledge and background in Databases Management, Neo4j, Natural Language Processing/Large Language Models are an asset.

Topic 2: Automated Planning and Scheduling for a Crewed Lunar Outpost
Experience in AI applications development and Python is needed. Knowledge and background in Space Systems, XAI, Natural Language Processing/Large Language Models are an asset.

Topic 3: Unlock the Future of Autonomous Space Systems with LLM Agents
Experience in AI applications development and related environments. Knowledge and background in embedded systems are an asset.

Topic 4: Implementation of a Semantic Segmentation demo in a Mars Environment
Experience in AI applications development, Python and Godot/Unity programming is needed.

Topic 5: Building a playable Mars Environment
Experience in Godot/Unity programming and 3D modelling in Blender is needed. Knowledge in scripting is an asset.

Topic 6: AI at the Edge in Space
Experience in Godot/Unity programming and 3D modelling in Blender is needed. Knowledge in scripting is an asset.