Research Engineer - Multimodal Language Models

About the Role

Key Responsibilities

• Design and develop large-scale annotation efforts for model post-training.
• Build tools to evaluate and benchmark multimodal language models.
• Develop large-scale AI training and inference methods.
• Ensure efficient implementation of models & systems for data processing and training.
• Build tools to visualize, evaluate and filter datasets.
• Collaborate with research and engineering teams across Luma to transfer research to products and services.
• Implement cutting-edge product prototypes based on multimodal generative AI.

Requirements

• Expertise in Python & Pytorch, including practical experience working with the full development pipeline from data processing, preparation & data loading to training and inference.
• Experience processing large-scale text data, or (bonus) interleaved data spanning audio, video, image, and/or text.
• Hands-on experience in developing or benchmarking at least one of the following topics: LLMs, Vision Language Models, Audio Language Models, generative video models.

Nice to Have

• Experience in design and development of annotation tools.
• Experience in synthetic data.

Qualifications

• No specific formal qualifications are explicitly mentioned in the description.

Benefits & Perks

• The pay offered to a successful candidate will be determined by various factors, including education, work experience, location, job responsibilities, certifications, and more.
• IntelliPro provides a comprehensive benefits package, all subject to eligibility.