Description

In this internship, you will collaborate with scientists to enhance internal LLM capabilities at Roche, focusing on fine-tuning models for drug discovery while developing training pipelines and evaluation protocols.

Requirements

• •Enrollment in a Master’s or PhD program in Computer Science, Computational Linguistics, or a related field
• •Strong coding skills in Python and PyTorch
• •Hands-on experience with LLM training or fine-tuning frameworks
• •Familiarity with RLHF/DPO workflows and human-in-the-loop alignment methods
• •Understanding of evaluation metrics for LLMs
• •Interest in applying AI to clinical and biomedical domains
• •Very good interpersonal and communication skills
• •Certificate from the university for Non-EU/EFTA citizens

Tasks

• •Collect and create datasets
• •Fine-tune open-source LLMs using curated datasets
• •Implement and iterate on training pipelines
• •Design evaluation protocols for accuracy and safety
• •Establish scalable training and evaluation workflows
• •Maintain experiment logs and model documentation

Tools & Technologies