easyTEM

Research Associate (f/m/d) for DFG-funded project about using AI Assistance for Interface Operation (easyTEM)
Humboldt University of Berlin | Faculty of Mathematics and Natural Sciences | Department of Computer Science
Full-time | Fixed-term until 31 March 2029
Location: Berlin-Adlershof (in-person position)
Reference: DR/030/26
Application Deadline: 02.03.26 (AoE)
Contact person for questions: thomas.kosch@hu-berlin.de

About the Role

In easyTEM, you will develop and evaluate AI agent “copilots” for complex visual user interfaces, systems that help users operate sophisticated software safely and effectively, supporting users without deep domain expertise. The project builds an intelligent software platform that makes interaction more effricient, safer to run, and easier to access by users without domain knowledge, with a strong emphasis on human-AI interaction, robust UI assistance, and trustworthy automation. You will work in an interdisciplinary team and collaborate closely with partners in physics and bioinformatics who frequently deal with visual complex interfaces. While your main focus remains on AI-powered interaction design and interface intelligence you will evaluate assistant capabilities with real-world users. The position is located at the Humboldt University of Berlin and will take place with the HCIstudio research group. The position is an in-person position.

Your Responsibilities

  • Conduct research within easyTEM to build a software platform that supports demanding workflows via intelligent assistance, safe interaction concepts, and remote access
  • Conceptualize, prototype, and evaluate interactive systems for complex GUI workflows
  • Develop and evaluate AI-assisted functions
  • Model and analyze user interaction on GUIs
  • Plan, run, and analyze user studies and usability tests
  • Contribute to open-source artifacts and publish scientific results
  • Opportunity to pursue a PhD as part of the position

Your Profile

  • Excellent Master’s degree (or equivalent) in Computer Science, Human-Computer Interaction, Data Science, or a related field
  • Experience in prototyping/software development and/or empirical evaluation methods
  • Very good English skills (at least C1)
  • Independent, structured, and collaborative working style
  • Experience with LLM integration and evaluation (e.g., prompting, tool-augmented workflows, RAG, fine-tuning)
  • Integarting LLM-based assistance in interfaces (e.g., Copilot, Chatbots)
  • Knowledge of machine learning for interaction data and/or computer vision (e.g., GUI understanding / parsing)

Nice to Have

  • Background in physics, bioinformatics, or experience with complex scientific application settings
  • Interest in interdisciplinary collaboration and scientific application contexts

What We Offer

  • Work at the cutting edge of AI agents that support users inside visual interfaces—from interface understanding to safe, assistive workflows
  • High creative freedom to explore interaction concepts that are trustworthy, transparent, and evaluation-driven
  • Applied, interdisciplinary environment with real users and constraints, including close collaboration with physics partners
  • Strong opportunities for publications, open-source outputs, and conducting a PhD thesis

How to apply

Please submit your application until 02.03.26 (AoE) (March 2nd 2026) and include:

  • cover letter
  • CV
  • relevant certificates/transcripts
  • publications (if applicable)
  • optional project links (e.g., GitHub)

Send your application in one PDF via email quoting DR/030/26 to thomas.kosch@hu-berlin.de. Other recipients or application modalities will not be accepted.

Equal Opportunity & Accessibility

HU Berlin is committed to equality and diversity. Applications from qualified women are particularly welcome. Applicants with severe disabilities will be given preference if equally qualified. Applications from people with a migration background are explicitly encouraged. Please submit copies only, as documents will not be returned.

Data protection information regarding the processing of personal data in the application and selection process is available on the website of the Humboldt University of Berlin.