Machine Learning Engineer, Gemini Factuality, CH at DeepMind

Google DeepMind is seeking a Machine Learning Engineer to enhance the quality and factual accuracy of the Gemini end-user product. The engineer will work on improving the accuracy of responses generated by Gemini models, both standalone and in Retrieval Augmented Generation (RAG) settings. The role involves data quality management, model building, quality evaluation, quality monitoring, and tuning the quality of RAG backends.

Key Responsibilities:

• Leverage model training strategies like Supervised Fine Tuning (SFT).
• Hillclimb on the Retrieval-augmented-generation (RAG) approaches.
• Collaborate with cross-functional and Research teams on applying state-of-the-art machine learning approaches to LLM-based products.

Requirements:

• 5 years of experience with software development.
• Bachelor’s degree or equivalent practical experience.
• 3 years of experience with state of the art GenAI techniques.
• 3 years of experience testing, maintaining, or launching software products.
• 1 year of experience with software design and architecture.
• 3 years of experience with ML infrastructure.

Advantage:

• PhD in Machine Learning or a related field.
• Experience with training and optimizing large language models.