Research engineer with a background in theoretical physics, applied AI / ML for physical and spatiotemporal systems, and scientific software engineering.

My work focuses on understanding complex natural and technical systems by connecting ideas across disciplines. I then use this understanding to model complex real-world processes and enable others to better grasp and use these systems.

Most repositories here are engineering-focused demonstrators: projects built to explore new technologies, validate ideas and create reusable reference implementations. I build them not only to deepen my own understanding, but also to make complex technical ideas transparent, readable and easier for others to understand and build upon.

Most of my projects explore topics such as

• Applied AI / Gen AI Workflow Systems Architecture and Large Language Models
• Scientific Machine Learning and Scientific Software Engineering
• Simulation and Surrogate Modelling

Rather than building applications around a particular framework, I enjoy understanding the systems behind them. Building software is my way of exploring complex ideas, testing my understanding and turning it into something tangible. I try to write code that remains clear, modular and easy to follow—not only because it is easier to maintain, but because well-structured software reflects a well-understood system.

Across projects I try to build software that is

• Modular rather than monolithic
• Understandable before optimized
• Easy to experiment with
• Well documented
• Reproducible
• Grounded in first-principles understanding

Whenever possible, I prefer to understand a technology deeply enough that I can build it—and build it clearly enough that I can explain it.

• PhD (Dr. rer. nat.) in Theoretical Physics
• 6+ years working in interdisciplinary applied AI / ML research projects collaborating with academia and industry
• Specializing in machine learning for spatiotemporal, physical and complex real-world systems
• University teaching, mentoring and project coordination

Python • PyTorch • NumPy • Pandas • Scikit-learn • Matplotlib • JavaScript • HTML • CSS

Gaussian Processes • Neural Networks • Statistical Modelling • Scientific Machine Learning • Surrogate Modelling

Retrieval-Augmented Generation (RAG) • Semantic Retrieval • Embedding Pipelines • Vector Search • Local LLM Integration

Modular Software Architecture • Scientific Software Development • Workflow and Systems design • API Integration • AI-assisted Software Engineering

I'm expanding my experience in modern AI engineering and AI workflow systems design, particularly

• Multimodal Retrieval-Augmented Generation
• Agentic AI, intelligent decision-making systems and intelligent tool use
• Local LLMs vs remote LLMs, prioritizing efficiency and security