Radiology ML Research Intern: 3D Vision & Foundation Models

StartX; Paris; Hybrid; EUR 13.000 - 22.000; Vollzeit

Benefits dieser Stelle

Premium computing hardware

Competitive stipend

Zusammenfassung

Raidium is seeking a research internship in France to contribute to foundation model pre-training and adaptation for radiology. You will join a team of ML researchers and radiologists to explore 3D vision, vision-language models, and clinical findings classification.

You will read literature, design and run experiments on 3D vision and VLMs, train models on a 130 TB dataset, benchmark on downstream tasks, collaborate with clinicians, and publish results where possible.

Qualifikationen

• Final year of MSc in Machine Learning, Computer Science, Mathematics or Electrical Engineering (for example MVA), or PhD student looking for a research internship
• Strong knowledge of Deep Learning, with Computer Vision and/or NLP
• Good Python skills, and experience with PyTorch

Aufgaben

• Read the latest literature on foundation models research for vision and language
• Design and run experiments on 3D vision and vision-language foundation models for radiology
• Train models at scale on Raidium's dataset of radiology exams and reports, using GPU clusters
• Benchmark your approach on clinical downstream tasks (detection, segmentation, classification, report generation)
• Work with radiologists to define the needs and to check the results
• Share your results with the team, and when possible, publish them in a top-tier conference or journal

Kenntnisse

Deep Learning

Computer Vision

NLP

Python

Ausbildung

MSc in Machine Learning / CS / Mathematics / Electrical Engineering

Tools

PyTorch

Jobbeschreibung

Raidium is seeking a research internship in France to contribute to foundation model pre-training and adaptation for radiology. You will join a team of ML researchers and radiologists to explore 3D vision, vision-language models, and clinical findings classification.

You will read literature, design and run experiments on 3D vision and VLMs, train models on a 130 TB dataset, benchmark on downstream tasks, collaborate with clinicians, and publish results where possible.