PhD Position in Foundation Model for Architectural Engineering

Challenge: Developing novel representation and foundation model for architectural engineering subjects.

Change: Advancing human-in-the-loop systems of architectural engineering knowledge tasks, with a focus on enhanced digital-human agency in design, engineering and construction coordination.

Impact: Enhancing adaptability and scalability of AI models in building engineering.

Job Description

We are seeking a highly motivated PhD candidate to pioneer the integration of architectural knowledge and logic directly into the latent manifold of Generative AI. This position operates at the cutting edge of AI and the Built Environment, challenging you to architect a native Multimodal Foundation Model for Architectural Engineering designed for "Constraint-to-BIM (Building Information Modelling)" synthesis. You will investigate how deep neural networks can internalize complex building dynamics (e.g., structural load paths, material thermodynamics) and regulatory logic (e.g., zoning envelopes) as native languages, facilitating rigorous, carbon-centric decision-making. If you are driven to transcend generic, task-agnostic generation and advance frontier AI toward structurally and functionally viable design, join us in defining this new scientific paradigm.

Your core research will target three novel frontiers:

• Heterogeneous input encoding: develop architectures to unify disparate modalities into a coherent embedding space.
• Causal chain-of-thought: engineer mechanisms for the model to manage long-horizon dependencies in construction sequencing and system-level planning.
• Constraint-aware latent reasoning: innovate tokenization strategies that semantically bind geometric primitives with performance constraints to ensure "construction-valid" generation.

You will apply parameter-efficient fine-tuning (e.g., low-rank adaptation) to an open-source Multimodal Foundation Model (MMLM) for domain-specific adaptation, combining it with novel pretraining objectives (e.g., topology-preserving loss, consistency-based self-supervision) to overcome engineering data scarcity and enable autonomous reasoning for tasks such as structural integrity, HVAC routing, and climate-adaptive design.

You will work within an interdisciplinary ecosystem of domain experts and stakeholders from industry and the public sector. A critical component of your work will be establishing rigorous evaluation protocols to probe the model's physics-consistency and logic-adherence, moving the field beyond standard perceptual metrics. Your contributions will define the state-of-the-art through publications in leading AI and built-environment venues, the release of open benchmarks for architectural reasoning, and the formulation of methodological guidelines for reliable domain-specific foundation models. Ultimately, your research will deliver a unified native foundation model, providing the multi-level disciplinary intelligence essential for the early-stage development of sustainable, high-performance buildings.

Requirements

• MSc in Computer Science, Artificial Intelligence, Data Science or Architectural Engineering (with strong computational focus), demonstrating a solid portfolio at the intersection of Deep Learning and geometric/physical data.
• Keen interest and/or experience in large-scale model architectures, specifically focusing on self-supervised learning, latent space alignment, or multimodal representation learning.
• Solid programming skills (e.g. Python) and familiarity with modern ML frameworks such as PyTorch or TensorFlow.
• Motivation to work with complex and heterogeneous AEC (Architecture, Engineering, and Construction) data.
• Hands-on experience in at least one of the following frontier areas: Geometric Deep Learning (GNNs, 3D-CNNs), Computer Vision, or Vision-Language Models (VLMs).
• Excellent scientific communication skills in English and a driven ambition to publish high-impact research in top-tier AI and Built Environment venues.
• Ability to bridge disciplines, translate complex problems into potential solutions, and collaborate effectively with domain experts and industry stakeholders.

TU Delft (Delft University of Technology)

Delft University of Technology is built on strong foundations. As creators of the world-famous Dutch waterworks and pioneers in biotech, TU Delft is a top international university combining science, engineering and design. It delivers world class results in education, research and innovation to address challenges in the areas of energy, climate, mobility, health and digital society. For generations, our engineers have proven to be entrepreneurial problem-solvers, both in business and in a social context.

Faculty Architecture & the Built Environment

The Faculty of Architecture and the Built Environment has a leading role in education and research worldwide. The driving force behind the faculty’s success is its robust research profile combined with the energy and creativity of its student body and academic community. It is buzzing with energy from early in the morning until late at night, with four thousand people studying, working, designing, conducting research and acquiring and disseminating knowledge. Our faculty has a strong focus on 'design-oriented research’, which has given it a top position in world rankings.

Staff and students are working to improve the built environment with the help of a broad set of disciplines, including architectural design, urban planning, building technology, social sciences, process management, and geo-information science. The faculty works closely with other faculties, universities, private parties, and the public sector, and has an extensive network in the Netherlands as well as internationally.

Conditions of Employment

Doctoral candidates will be offered a 4-year period of employment in principle, but in the form of 2 employment contracts. An initial 1.5 year contract with an official go/no go progress assessment within 15 months, followed by an additional contract for the remaining 2.5 years assuming everything goes well and performance requirements are met.

Salary and benefits are in accordance with the Collective Labour Agreement for Dutch Universities, increasing from €2770 per month in the first year to €3539 in the fourth year. As a PhD candidate you will be enrolled in the TU Delft Graduate School. The TU Delft Graduate School provides an inspiring research environment with an excellent team of supervisors, academic staff and a mentor. The Doctoral Education Programme is aimed at developing your transferable, discipline-related and research skills.

The TU Delft offers a customisable compensation package, discounts on health insurance, and a monthly work costs contribution. Flexible work schedules can be arranged.

For international applicants, TU Delft has the Coming to Delft Service. This service provides information for new international employees to help you prepare the relocation and to settle in the Netherlands. The Coming to Delft Service offers a Dual Career Programme for partners and they organise events to expand your (social) network.

Additional Information

If you would like more information about this role, please contact Dr ir. Pei-Yu Wu, Assistant Professor, via +31 (0)634115402 or p.y.wu@tudelft.nl or Dr ir. Michela Turrin, Associate Professor, via +31 (0)62921839 or m.turrin@tudelft.nl.