CV
A compact overview of research direction, engineering practice, and research engineering experience.
Research Profile
Research direction around reliable AI systems, evaluation, certified robustness, randomized smoothing, data-centric and software-engineering questions for AI, and the practical limits that appear when systems enter real workflows.
Working Style
I am drawn to problems where the goal is not fully specified yet but still needs precise investigation. My strength is translating ambiguous goals into testable hypotheses, reproducible experiments, maintainable tools, and communication that lets research and engineering teams make decisions.
Research Output
Master's thesis work on certified robustness and training efficiency, including benchmarking and interventions such as LossProj and Selective Backpropagation. The broader thread is controlled evaluation: exposing assumptions, measuring trade-offs, and keeping claims tied to what the evidence supports.
Engineering And Research Practice
At the August-Wilhelm Scheer Institute, I work on research-facing AI and cloud software infrastructure, including API management and mocking with an agentic LLM-based development assistant, multi-provider cloud cost analysis with Odoo integration, and federated learning infrastructure. Earlier, at EnigmaAI, I led backend and ML-related development for a virtual assistant SaaS connecting NLP-powered chatbot workflows with social messaging channels.
Technical Breadth
I have accumulated working experience with a diverse set of technologies: Python, C#, TypeScript, C++, Java, SQL, PyTorch, LangGraph, scikit-learn, NumPy, SciPy, Pandas, ASP.NET, Node.js, Express, FastAPI, Angular, React, Flutter, Azure, AWS Lambda/ECS, Docker, Odoo, Linux, MongoDB, PostgreSQL, and Git.