Stage : Construction automatique de modèles de processus hospitaliers multi- flux par grands modèles de langage (LLM)
Core
Design and prototype an LLM-based pipeline to automatically construct hospital process models (BPMN, process graphs) from both historical logs and predicted patient pathways to optimize emergency department workflows.
Role type
Research intern (M2/Engineering school final year) in AI and Process Mining
Builds
Automated process modeling pipeline integrating retrospective logs and prospective LLM predictions
Domain
Healthcare operations / Artificial Intelligence / Process Mining
Deliverable
production ML models
Required skills
Python, Large Language Models (LLM), Process Mining (pm4py)
Preferred skills
Fine-tuning, Prompting, API integration, Event log normalization
Technologies
Python, LLMs, pm4py, SIH logs
Responsibilities
Develop a pipeline to ingest normalized event logs and serialized LLM-predicted patient paths; Generate as-is and to-be process models; Compare models to detect bottlenecks and support decision-making; Prototype counterfactual simulations.