About Me

I'm an AI engineer interested in how machines can understand the world and act within it. My work focuses on building multimodal models that connect perception and action — particularly vision-language and vision-action systems for robotic applications.

I enjoy working on problems that go beyond model training. Much of my work involves building the full pipeline around machine learning systems: training large multimodal models, designing efficient inference infrastructure using tools like Python and Rust, and integrating learned policies into real-time environments and simulations.

Recently, I've been working on vision-language models and exploring how they can be extended into vision-action models that allow robots to interpret scenes and generate meaningful actions. I'm particularly interested in scalable training workflows, real-time inference architectures, multimodal reasoning, and building robust ML systems that operate in real-world environments.

My goal is to build AI systems that bridge perception, reasoning, and action in real-world robotic settings. My background spans machine learning, distributed systems, and high-performance programming, and I enjoy bridging research ideas with practical engineering.

Background

My journey started in mechanical engineering, pivoted through industrial automation at Amara Raja, and found its home in deep learning and robotics at OVGU Magdeburg. Over four years at Agile Robots, I've grown from a software developer to leading the Foundation Models team, driving the technical strategy for next-generation robotic intelligence.

Education

Skills

Certifications

Languages