Projects

Master's thesis at Volkswagen AG: an LLM web agent built on agentic context engineering (structured DOM preprocessing), knowledge-graph agent memory, cost-aware routing to vision-language models, and human-in-the-loop guardrails. Strict success rate of 69.7% to 95.9% across multiple benchmarks. Related paper: WCI — The Web Context Interface (CIKM 2026).

Neuro-symbolic dashboard for sustainable and dynamic exploration of research knowledge. Technical team lead of 5 students. Received the Best Posters and Tools Award at REFSQ 2026.

Bremen Big Data Challenge 2026: winning solution (1st place, March 2026). Machine learning pipeline for age-related disease detection from multimodal data.

Neuro-symbolic dashboard for exploring Open Research Knowledge Graph (ORKG) templates. Demo built around the AIKG-SD 2026 summer school.

Research framework for LLM-guided curriculum learning in multi-agent reinforcement learning (Blackjack). Code for the DAI 2025 oral paper.

Automation pipeline that turns NLP4RE PDF ID cards into structured data in the Open Research Knowledge Graph (ORKG) to answer competency questions over the knowledge-graph data.

A comprehensive, modular and robust Python toolkit for ontology alignment. Best Resource Paper Award at ESWC 2025.

LLM-based assistant that supports researchers across the entire research and writing process with reliable, source-grounded answers.

Automated Machine Learning Agent implementation.

Bachelor's thesis (grade 1.0) at Samian Soft: optimized visitor and freight distribution by combining routing algorithms (VROOM) with clustering methods, for shorter routes and higher efficiency.

Master course repository at Leibniz University Hannover.

Project focused on sentiment analysis and segmentation.

Review and implementation of Neural Networks and Generative Networks.