AI Research Intern, TAO Multi-Modal Model Development - 2026
Core
Extending an internship to develop and fine-tune multi-modal AI models, specifically vision-language models and universal segmentation systems, using NVIDIA's TAO framework.
Role type
AI Research Intern (Multi-Modal Model Development)
Builds
Scalable, high-impact AI solutions for tasks like image segmentation and cross-modal understanding.
Domain
Artificial Intelligence / Deep Learning / Computer Vision
Deliverable
production ML models
Required skills
Python, PyTorch, neural network architectures, transformers, multi-modal learning techniques, model development, experiment design, benchmarking
Preferred skills
vision-language models, image segmentation, large-scale pretraining
Responsibilities
Develop and fine-tune multi-modal AI models using NVIDIA's TAO Toolkit; Contribute to the design and implementation of vision-language models and universal segmentation systems; Conduct experiments and benchmarking to evaluate model accuracy, robustness, and scalability; Collaborate with cross-functional teams to integrate research into production-level pipelines; Participate in research discussions, code reviews, and technical documentation.
