Industry internship at Det Norske Veritas (DNV) in the healthcare team of Group Research and Development. Funded by Digital Life Norway (DLN). Oslo. 1 September- 1 December 2023.
Generation of synthetic tabular data in healthcare using large language models (LLMs) like GPT (from Hugging Face).
Evaluation of the synthetic data employing a wide variety of similarity and privacy metrics (library SDMetrics) as well as downstream machine learning models (library Pytorch).
Use of other well-established machine learning models to generate tabular data (like CTGAN, library SDV).
Production-level code with Kedro.
🧐 Press article | ☕ Blog about my experience at DNV