Research Associate (E13, DIMA, IV-256/25)

Research Assistant (E13, DIMA)

The Berlin Institute for the Foundations of Learning and Data (BIFOLD) at TU Berlin is seeking a Research Associate (f/m/d) in the field of Machine Learning and Computer Vision for an Agility subproject. The project is part of the NebulaStream team, led by Prof. Volker Markl and Dr. Steffen Zeuch. Entitled "Analysis of Intraoperative Sensor Data Streams to Support Minimally Invasive Surgical Procedures", the project aims to perform an integrated analysis of sensor data streams captured during minimally invasive surgeries. To this end, complementary sensor data will be combined using the NebulaStream framework, enabling accurate detection and analysis of the various surgical phases. These insights are intended to support risk assessment, clinical decision-making, and logistics. For the analysis of echocardiographic and endoscopic video sequences, the project will evaluate scalable, resource-efficient, and clinically applicable multi-task deep learning networks.

Tasks

Independent and responsible research in the field of Machine Learning and Computer Vision.

Processing of image and sensor data streams may involve the application of multi-task deep learning methods. Alternative approaches for data analysis, data fusion, and appropriate visualizations are to be autonomously designed and evaluated. Key factors to be considered in the conception and evaluation of suitable solutions include the hardware constraints and real-time requirements of the clinical environment.

Requirements

• Successfully completed university degree (Master, Diplom or equivalent) in computer science, industrial engineering, or business informatics.
• Experience in image and sensor data processing and machine learning required, preferably in the area of neural network architectures for geometric and high-dimensional data
• Excellent programming skills in Python, NumPy/SciPy, PyTorch/TensorFlow, and C++ are essential
• Good knowledge of German and/or English required; willingness to acquire the respective missing language skills

Desirable:

• Experience in the analysis of medical image data is of advantage.

Employer: TU Berlin / BIFOLD
Salary Grade: E13, TVL Berliner Hochschulen
Starting Date: at the earliest possible (for 3 years)
Closing Date: July 25, 2025
Full job posting: IV-256/25