The European Space Agency (ESA) is looking for a Machine Learning Development intern in its CAVES and PANGAEA team at the European Astronaut Centre (EAC) in Porz-Wahn, Germany, to help improve the software that recognises minerals and rocks for astronaut geology training. Applications close on 30 November 2026 at 23:59 CET/CEST. At Youth Works Hub, we reviewed the full vacancy and set out every task, skill and eligibility rule below.
Vacancy Details: Requisition 21138
The vacancy sits within ESA's Human and Robotic Exploration Directorate. These are the formal details:
๐ Job Requisition ID: 21138
๐ Type of appointment: Intern
๐๏ธ Directorate: Human and Robotic Exploration
๐ Workplace: Porz-Wahn, DE
๐บ๏ธ Location: EAC, Porz-Wahn, Germany
๐ฃ Publication method: External Only
๐ Closing date: 30 November 2026, 23:59 CET/CEST
Meet the CAVES and PANGAEA Team
The CAVES and PANGAEA team specialises in training programmes that equip astronauts and mission developers with scientific, expeditionary and behavioural skills. Complementary to their training goals, the programmes are used as research and development platforms to advance several of ESA's technological, scientific and operational areas.
Underground and Field Training: CAVES and PANGAEA
The group's primary output is focused around two training programmes:
๐ณ๏ธ CAVES: a course that uses natural cave systems for expeditionary and human behavioural and performance training
๐ PANGAEA: a course for geological and astrobiological field training
Inside the PANGAEA Analytical Toolset
One tool developed for PANGAEA and future missions is the PANGAEA Analytical Toolset. It comprises two parts:
๐๏ธ The Mineralogical Database (MinDB): a curated collection of information on all the minerals found on the Moon, Mars and in meteorites
๐ค The Machine Learning (ML) software: it combines deep learning multi-class and multi-label classification algorithms analysing fused multi-modal spectroscopic measurements, enabling a significant accuracy increase of mineral and rock recognition
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
๐ All opportunities in Germany โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
Topic of the Internship
The vacancy names four areas for this internship:
๐ง Machine Learning
๐ป Software Development
๐ก Spectroscopic Analysis
๐ช Planetary Mineralogical Characterisation
Choose Your Stream: Three Complementary Paths
Interns are sought to contribute to the ongoing enhancement, integration and deployment of the PANGAEA analytical toolset. Depending on the primary specialization, the intern will join one of three complementary streams:
โ๏ธ Machine Learning (ML) Development and Operations
๐งฉ Applied Software Development
๐ฌ Mineralogical & Spectroscopic Data Curation
What You Will Work On
The activities will include:
๐ Developing and testing ML classifiers for multi-method spectroscopic data, including signal unmixing, multi-modal data fusion, hierarchical classification, incremental few-shot learning, conformal prediction and uncertainty estimation
๐ ๏ธ Building, maintaining and optimising the PANGAEA Analytical Toolset software applications
โ Evaluating ML outputs against ground-truth multi-spectral MinDB reference archives and petrographic databases, and expanding and curating the datasets
Start Date, Duration and Cover Letter
ESA is looking for internships to start throughout 2027. Candidates are invited to state their preferred starting date in the cover letter.
According to ESA's announcement of its 2027 student internships, an internship lasts between 3 and 6 months and starts from February 2027. The exact start date is agreed with the ESA tutor.
Who Can Apply: University Enrolment
You must be a university student, preferably in your final or second-to-last year of university course at Master's level. In addition, you must be able to prove that you will be enrolled at your university for the entire duration of the internship.
Which Nationalities Are Accepted
Applications are only considered from nationals of one of the following States: Austria, Belgium, Cyprus, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, the Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland, and the United Kingdom.
Working Languages
The working languages of the Agency are English and French. A good knowledge of one of these is required. Knowledge of another ESA Member State language is an asset.
Machine Learning Experience ESA Expects
Candidates should possess practical experience in classification techniques, particularly those based on Machine Learning supervised hierarchical classification and regression. Relevant experience includes, but is not limited to:
๐๏ธ Data mining
๐ Data fusion
๐ Statistics
๐ฏ Clustering
ใฐ๏ธ Signal decomposition/unmixing
๐ Incremental few-shot learning
๐งฎ Conformal prediction
๐ธ๏ธ CNN or KAN ML architectures
Practical experience in processing and analysing data derived from analytical instrumentation, or working with databases, is also valuable.
Python and Scientific Computing Skills
Academic or professional proficiency with the Python programming language and the core scientific computing and ML libraries is expected: NumPy, JAX, TensorFlow, Keras, CatBoost and Scikit-learn.
Extra Skills That Count as an Asset
Additional skills and experience in one or more of the following key areas will be considered a valuable asset:
๐ Accelerating Python code using Cython, JAX or similar performance optimisation tools
๐ผ๏ธ Practical experience with analysis and visualisation of scientific data (Matplotlib)
๐ง Familiarity with MLOps workflows and dataset integration
๐ For applied ML, experience with modern web application development, user interface design for applications and web frontend integration, and backend services using FastAPI
Behavioural Competencies Assessed
The vacancy lists six ESA core behavioural competencies, described in the ESA Core Behavioural Competencies guidebook:
๐ Result Orientation
โก Operational Efficiency
๐ค Fostering Cooperation
๐ฌ Relationship Management
โป๏ธ Continuous Improvement
๐ญ Forward Thinking
How to Apply
Candidates are invited to state their preferred starting date in the cover letter. During the interview, your motivation for applying to this role will be explored.
Selection Tests and Export Control Rules
During the recruitment process, the Agency may request applicants to undergo selection tests. Applicants must be eligible to access information, technology, and hardware which is subject to European or US export control and sanctions regulations.
Applying to More Than One ESA Internship
According to ESA's announcement of its 2027 student internships, one candidate can apply for a maximum of 2 vacancies.
Equal Opportunities and Accessibility
ESA is an equal opportunity employer, committed to achieving diversity within the workforce and creating an inclusive working environment. Applications are welcome from all qualified candidates irrespective of gender, sexual orientation, ethnicity, religious beliefs, age, disability or other characteristics.
ESA values diversity and welcomes people with disabilities. Whenever possible, it seeks to accommodate individuals with disabilities by providing the necessary support at the workplace. The Human Resources Department can also provide assistance during the recruitment process.
Good to Know Before You Apply
The information published on ESA's careers website regarding internship conditions is correct at the time of publication. It is not intended to be exhaustive and may not address all questions you would have. The vacancy does not state a stipend or allowance.
Looking for more roles like this one? Find them in the internships and traineeships section of Youth Works Hub.
Frequently Asked Questions
Where is the Machine Learning Development internship based?
The internship is based at EAC in Porz-Wahn, Germany, within ESA's Human and Robotic Exploration Directorate. The vacancy lists Porz-Wahn, DE as the workplace.
What is the deadline to apply?
The closing date is 30 November 2026 at 23:59 CET/CEST. Applications are submitted through the ESA careers website.
What will I do as a Machine Learning Development intern?
You will contribute to the enhancement, integration and deployment of the PANGAEA analytical toolset. This includes developing and testing ML classifiers for spectroscopic data, building and maintaining the toolset software, and evaluating ML outputs against MinDB reference archives and petrographic databases.
What are the three streams of this internship?
Depending on your primary specialization, you join one of three complementary streams: Machine Learning (ML) Development and Operations, Applied Software Development, or Mineralogical & Spectroscopic Data Curation.
Who can apply for this internship?
You must be a university student, preferably in your final or second-to-last year of university course at Master's level. You must also be able to prove that you will be enrolled at your university for the entire duration of the internship.
Which nationalities are eligible?
Applications are only considered from nationals of ESA's listed States, which include Germany, France, Italy, Spain, the United Kingdom and 22 others named in the vacancy.
When does the internship start and how long does it last?
The vacancy looks for internships to start throughout 2027, and candidates state their preferred starting date in the cover letter. According to ESA's 2027 student internships announcement, an internship lasts between 3 and 6 months and starts from February 2027, with the exact start date agreed with the ESA tutor.
Which languages do I need?
The working languages of the Agency are English and French, and a good knowledge of one of them is required. Knowledge of another ESA Member State language is an asset.
Which programming skills are required?
The vacancy asks for academic or professional proficiency with Python and the core scientific computing and ML libraries NumPy, JAX, TensorFlow, Keras, CatBoost and Scikit-learn. Cython, Matplotlib, MLOps workflows and FastAPI are listed as valuable additional skills.
Which machine learning experience matters most?
Candidates should have practical experience in classification techniques, particularly supervised hierarchical classification and regression. Data mining, data fusion, statistics, clustering, signal unmixing, incremental few-shot learning, conformal prediction and CNN or KAN architectures are named as relevant experience.
How many ESA internship vacancies can I apply for?
One candidate can apply for a maximum of 2 vacancies, according to ESA's announcement of its 2027 student internships.
Is the internship paid?
The vacancy does not state a stipend or allowance.
What happens during the selection process?
Your motivation for applying to the role will be explored during the interview. ESA may also ask applicants to undergo selection tests, and applicants must be eligible to access information, technology and hardware subject to European or US export control and sanctions regulations.
Deadline
โณ 30 November 2026
Links
๐ Full vacancy and application (Apply now): https://jobs.esa.int/job/Porz-Wahn-Intern-in-the-CAVES-&-PANGAEA-Team-Machine-Learning-Development/1453000933/
๐ Detailed information on this internship position: https://www.esa.int/About_Us/EAC/Space_training_team_Planetary_Mineral_Database_Development_and_Validation_of_Spectra_Classification_Methods
๐งญ What is PANGAEA: https://www.esa.int/Science_Exploration/Human_and_Robotic_Exploration/CAVES_and_Pangaea/What_is_PANGAEA
๐ง What is CAVES: https://www.esa.int/Science_Exploration/Human_and_Robotic_Exploration/CAVES_and_Pangaea/What_is_CAVES
๐ ESA Student Internships 2027 announcement: https://www.esa.int/About_Us/Careers_at_ESA/ESA_Student_Internships_2027_are_now_open
๐ ESA website: http://www.esa.int
๐ง ESA student internships contact (diversity and accessibility questions): ESA.Student.Internships@esa.int
