This internship focuses on supporting the development of an AI-powered system designed to detect manipulated images and figures in academic publications and integrate it into editorial workflows. You will research and apply computer vision and image forensics techniques while assisting in building and evaluating machine learning and deep learning models to identify anomalies in scientific figures. The role offers hands-on experience collaborating with technical teams to manage datasets, improve model performance, and contribute to reliable image integrity detection solutions.

As part of our commitment to fostering a collaborative team environment, this role requires working fully on-site.

Tasks & Responsibilities

• Support the preparation of a proof-of-concept prototype capable of detecting image manipulations and its integration into the editorial workflows.
• Contribute to the development of an AI-driven system designed to detect various types of image and figure manipulations in academic publications.
• Conduct research and experimentation on computer vision and image forensics techniques, including duplication detection, splicing, retouching, and contrast or brightness alterations.
• Develop and evaluate machine learning and deep learning models (e.g., CNNs, autoencoders, transformers) for anomaly and similarity detection in scientific figures.
• Collect, preprocess and analyze image datasets extracted from academic manuscripts, ensuring data quality and representative coverage of manipulation types.
  
• Collaborate with data engineers and data scientists to design scalable training pipelines, manage large image datasets, and benchmark algorithmic performance.
  
• Explore and implement relevant open-source tools, datasets, and research methods in the field of digital image forensics.
  
• Document experimental results, model architectures, and findings, contributing to internal technical reports and presentations.

Requirements

• Master’s degree in Data Science, Machine Learning, Artificial Intelligence, or a closely related field is required; a PhD in these areas is considered a strong advantage.
  
• Less than 2 years of professional or hands-on experience in Computer Vision, AI/Data Science, and Linux-based environments.
  
• Advanced proficiency in English, with strong written and verbal communication skills.
  
• Working knowledge of Microsoft O365 tools for documentation and collaboration.
  
• Basic proficiency in Python for building complex applications and developing machine learning models.
  
• Familiarity with technologies such as FastAPI, Celery, and Keycloak.
  
• Solid understanding of core Artificial Intelligence principles and methodologies.
  
• Hands-on exposure to deep learning frameworks such as PyTorch or TensorFlow for image analysis and model development.
  
• Experience using computer vision libraries including OpenCV, scikit-image, and Pillow for image preprocessing and feature extraction.
  
• Understanding of convolutional neural networks (CNNs) and modern vision architectures such as ResNet, Vision Transformers, and autoencoders.
  
• Foundational knowledge of image forensics and manipulation detection techniques, including error level analysis (ELA), copy-move detection, resampling analysis, and metadata inspection.
  
• Familiarity with image similarity and retrieval techniques such as feature embeddings and perceptual hashing.
  
• Basic exposure to containerization, DevOps, and MLOps tools including Docker, Kubernetes, Helm, Ansible, MLflow, and Kubeflow.
  
• Fundamental understanding of version control systems and CI/CD pipelines using Git.
  
• Strong problem-solving and analytical thinking skills.
  
• Excellent communication abilities with the capability to explain complex technical concepts to non-technical stakeholders.
  
• Ability to work effectively both independently and within a collaborative team environment.
  
• Comfortable operating in fast-paced, dynamic settings.
  
• Experience working within Scrum / Agile workflows, with a collaborative mindset and leadership potential.
  

What we offer

• The opportunity to contribute to the academic/scientific community;
• Flexible working hours;
• Team bond strengthening through team-building events;
• Professional growth opportunities with our global training system;
• Working in a collaborative and socially responsible team;
• Company retreat facility;
• Full-coverage insurance for accidents/daily sickness;
• Prime location near Basel train station and city center;
• And more

If you are interested in this position, we look forward to receiving:

• A motivational letter (EN) that briefly describes your motivations for joining MDPI
• A resume (EN) including personal information, past & current education,
• Copy of ID and Diplomas
• If available: reference letters and certificates.

About MDPI

A pioneer in scholarly open access publishing, MDPI has supported academic communities since 1996. Our mission is to foster open scientific exchange in all forms, across all disciplines. We operate more than 400 diverse, peer-reviewed, open access journals supported by over 66,000 academic editors. We serve scholars from around the world to ensure the latest research is openly and broadly available.

MDPI is headquartered in Switzerland with additional offices in Europe, Asia and North America. We are committed to ensuring that high quality research is made available as quickly as possible. We also support sustainability projects, with sustainability as a key theme in many journals and through the MDPI Sustainability Foundation.

Initiatives

At MDPI, we develop and maintain various platforms in order to better serve the scientific community. Please find here-below a list of our main platforms:

https://www.mdpi.com
https://www.mdpi.com/books/
https://sciprofiles.com
https://sciforum.net
https://www.scilit.net
https://www.preprints.org
https://encyclopedia.pub

#LI-Onsite