Philipp Langen Email & Phone Number

Dresden, Sachsen, Deutschland

- Developed a document processing pipeline for the construction sector to automatically structure and extract information from project documentation to support certification processes (CO2,LCA) using large language models (Azure OpenAI supported by RAG via weaviate).- Developed several ML-based microservices (FastAPI) for an end-to-end invoice processing pipeline.- Trained, evaluated and improved deep learning models for document classification, optical character recognition (OCR) and key information extraction (KIE) (Pytorch, scikit-learn).- Temporarily took on additional responsibilities as product owner. Project coordination, backlog management and communication with technical and non-technical stakeholders.

Implemented and tested the Brown clustering algorithm as a means of dealing with rare words when training hybrid grammars for the induction of discontinuous phrase structures.

Master, Informatics, 1,1

Bachelor'S Degree, Computer Science