Blätter-Navigation

Offer 2 out of 29 from 19/05/26, 17:07

logo

Uni­versität der Kün­ste - Department of Building Planning & engineering (VPT) at the Institute of Architecture and Urban Planning

The Universität der Künste Berlin, situated in Berlin, Germany, is the largest art school in Europe. It is a public art and design school, and one of the four research universities in the city. The university is known for being one of the biggest and most diversified universities of the arts worldwide.

The rese­arch work at the Depart­ment of Sup­ply Plan­ning and Sup­ply Engi­nee­ring focu­ses on the deve­lop­ment of simu­la­tion-based methods, models and tools for energy-effi­ci­ent buil­ding design, for the ana­ly­sis of the buil­ding's phy­si­cal beha­vior and for the ener­ge­tic-func­tio­nal beha­vior of the buil­ding ser­vices. Through the rese­arch work, the tools for ener­ge­tic buil­ding and sys­tem simu­la­tion are to be bet­ter inte­gra­ted into the design and plan­ning pro­ces­ses of the buil­ding indus­try and made easier to use.

Research internship and/or thesis in the field of Building Physics and Machine
Learning

The Universität der Künste Berlin, VPT department, invites you to contribute to the FACaiDE
(AI-supported analysis of energy efficiency of facades) project (link to the project).

The building sector plays a central role in the energy transition, yet detailed façade information
required for accurate building energy assessment is often missing or incomplete. Recent advances in AI and computer vision enable the automated extraction of building-related parameters directly from street-level images and video data.

Here comes the project FACaiDE, where the goal is to develop a mobile, edge-sensing device that
performs real-time building analysis using hybrid AI and multi-modal sensing. We are looking for
motivated students to take on the following thesis tracks for 2026:

Tasks:

The thesis explores a non-invasive method for estimating building façade energy efficiency
parameters by combining infrared thermography, environmental sensor data, and simplified machine
learning models. The aim is to reduce measurement time compared to conventional steady-state
methods while maintaining acceptable accuracy. Both physics-based approximations and ML-based regression approaches are evaluated using transient thermal data collected under real outdoor conditions.

Requirements:

  • Bachelor's degree from the 4th semester onwards or Master's degree in a field such as energy
    and building technology, computer science, or a comparable subject.
  • Background in Python and Machine Learning or willing to learn.
  • Interest in implementation of theoretical physics in real world scenario.
  • Basic knowledge of energy for buildings is a plus.

What we offer:

  • International, inclusive and collaborative team.
  • Active support in publishing findings in international journals and conferences.
  • Flexible start date (immediately or by arrangement) and working hours.

How to apply:

Please submit your application, quoting "FACaiDE01 Student Position", along with the usual documents
to Prof. Dr. Christoph Nytsch-Geusen via email (in a single PDF document) to nytsch@udk-berlin.de.
The Berlin University of the Arts (UdK Berlin) is committed to an equal opportunity and discrimination-free learning, teaching, and working environment and works to dismantle structural barriers (such as physical, linguistic, racial, age-related, gender-specific, heteronormative, and others). It aims to increase the proportion of women by hiring and promoting qualified women, particularly in leadership positions and in areas where they are underrepresented, with special consideration given to an intersectional approach. The UdK Berlin explicitly encourages qualified individuals with a migration background, Black people, and/or People of Color to apply. Applicants with a recognized severe disability will be given preference if equally qualified. Please indicate any severe disability in your application.

By submitting your application, you consent to your data being processed and stored electronically.