Google is seeking a Research Scientist specializing in Multilingual Large Language Models to join their Technology & Society organization. The role involves developing advanced methodologies for multilingual environments, focusing on pre-training multilingual models, enhancing multilingual instruction-tuning datasets, refining evaluation processes, boosting knowledge transfer across languages, and optimizing multilingual tokenization.
Key responsibilities include:
- Authoring research papers to share findings with the team and research community
- Researching and developing technology for improving multilingual Large Language Models (LLM)
- Developing pre-training techniques for LLMs in non-English target languages
- Collaborating with other research teams to expand multilingual LLM technology
- Working with Google's first-party partner teams to implement new multilingual technologies in production
The ideal candidate should have:
- PhD in Computer Science or a related field, or equivalent practical experience
- Coding experience in Python, JavaScript, R, Java, or C++
- Scientific publication submissions for conferences, journals, or public repositories
- Experience with modern Large Language Models and generative models
- Expertise in multilingual LLMs
- Recent publications in Generative Artificial Intelligence fields
This role offers the opportunity to work on cutting-edge research in multilingual AI, contribute to Google's global impact, and shape the future of language technology. The position is based in Tel Aviv, Israel, and requires English proficiency for effective global collaboration.