Curriculum Vitae
I build efficient NLP systems and applied machine learning products, focusing on deployable models, retrieval workflows, and explainable pipelines.
- Deployable LLM and encoder pipelines (retrieval, contract comparison, ticket automation).
- Research-to-production focus: distillation, efficiency benchmarking, explainability.
- Comfortable across Java, Python and TypeScript stacks with end-to-end delivery.
Profiles
Education
Swiss-European Mobility Program (SEMP), Computer Science
2024 – 2025École Polytechnique Fédérale de Lausanne (EPFL) · Lausanne, Switzerland
- Machine Learning, Advanced Probability, and Applied Data Analytics.
- Master thesis: Tiny Language Models for NLP Pipelines (with Prof. Robert West).
M.Sc. Computer Science
2023 – PresentKarlsruhe Institute of Technology (KIT) · Karlsruhe, Germany
- Focus: AI, Data Science, and IT Security; current grade 1.3.
- Research: explainable AI for micro-expression recognition.
B.Sc. Information Systems
2019 – 2023Karlsruhe Institute of Technology (KIT) · Karlsruhe, Germany
- Specialisation in Software Engineering, Data Science, and Finance.
- Thesis: Hidden Outliers in Manifolds (grade 1.0).
- Exchange semester: Budapest University of Technology and Economics (BME), 2022 – 2023.
Experience
AI Engineer (Working Student)
2024 – 2025dreifach.ai · Remote / Cologne, Germany
Built LLM-backed tools for insurance workflows (document analysis, internal chat, retrieval).
Implemented embedding-based contract comparison, automated ticket tagging, and privacy-aware archival pipelines.
Data Science Tutor
2024Karlsruhe Institute of Technology · Karlsruhe, Germany
Supervised Data Science Lab projects; grading, code reviews, and project guidance.
Software Engineer (Working Student)
2021 – 2023Vector Informatik GmbH · Karlsruhe, Germany
Developed Java tooling (Maven, SVN, JUnit, Mockito) for the PREEvision CASE suite.
Designed and shipped a plugin for the propagation rule framework; presented to the department.
Selected ML Systems & Research
TiME: Tiny Monolingual Encoders for Efficient NLP Pipelines
2025EPFL Datalab · Lausanne, Switzerland
- Distilled MiniLMv2-based encoders across 16 languages (three model sizes each) for deployable NLP pipelines.
- Evaluated on POS, lemmatization, dependency parsing, NER, and QA; measured latency, throughput, and energy per sample.
- Released models and code; presented at RethinkingAI@NeurIPS 2025 (Copenhagen).
Explainable NLP for ADHD Anamnesis
2025 – PresentKarlsruhe Institute of Technology · Karlsruhe, Germany
- Building an explainable NLP system for German primary school reports to support retrospective ADHD assessment, by training custom models and comparing them with API. Custom model training included sft of Encoder Models, and SFT + RL of Qwen3-4B.
- Python backend (models, REST API) plus React/Node.js frontend; evidence highlighting and clinical metrics (AUC/ROC, sensitivity, specificity).
Handwriting Synthesis Pipeline for Dataset Generation
2024Swiss AI Center & ETH Zurich · Zurich, Switzerland
Generated realistic handwritten images from LaTeX math exercises to stress-test OCR systems and produce multilingual training data.
Hidden Outliers in Manifolds
2023Karlsruhe Institute of Technology · Karlsruhe, Germany
Developed methods to generate and detect hidden outliers in high-dimensional data using autoencoders.
Explainable AI Lab
2023 – 2024Karlsruhe Institute of Technology · Karlsruhe, Germany
Implemented and compared explainable approaches for facial micro-expression recognition (prototype and evaluation).
Skills
Languages/Tools
AI/ML
Data
Languages
Leadership & Service
Resident Speaker
2020 – 2023Hadiko Student Dormitory · Karlsruhe, Germany
Represented 100+ residents, moderated weekly meetings, and coordinated community initiatives.