Research Scientist - Data Excellence for Evaluating Responsibly - Fixed Term Contract (12 month)

At Google DeepMind, we value diversity of experience, knowledge, backgrounds and perspectives and harness these qualities to create extraordinary impact. We are committed to equal employment opportunities regardless of sex, race, religion or belief, ethnic or national origin, disability, age, citizenship, marital, domestic or civil partnership status, sexual orientation, gender identity, pregnancy, or related condition (including breastfeeding) or any other basis as protected by applicable law. If you have a disability or additional need that requires accommodation, please do not hesitate to let us know.

Snapshot: 

Our team is responsible for data driven approaches to responsible AI specifically focusing on integrating diverse human values and viewpoints via datasets and metrics into the training and evaluation of Large Language Models responsibly.

As a Research Scientist, you will design, implement, and empirically validate inclusive automated approaches to data collection, data insights extraction and demonstrating impact on the reliability of evaluation and training of AI systems. The focus will be expanding existing textual approaches to multi-modal and multi-lingual data and AI capabilities.

About us: 

Our interdisciplinary team drives research in responsible data for evaluation of safety and fairness of AI systems, including identifying potential harms from current and future AI systems, and conducting (socio)technical research to mitigate them. 

The role:

Research Scientists spearhead innovative technical approaches, collaborating closely with internal AI research teams (like Responsibility, AI Safety and Foundational Research), product teams (including Gemini, Gemma), and external research partners.

Key responsibilities:

• Design and implement technical research approaches / algorithms for automatic insights extraction and statistically rigorous analysis of diverse data collections, integrating diverse viewpoints and values in safety evaluations and finetuning datasets, specifically focusing on expanding multi-modal capabilities beyond text.  The work will be done in coordination with the team’s broader technical agenda.
• Identify signals in data that correlate with possible risks and harms of foundation models and systemic impacts.
• Integrate solutions in existing infrastructure to accelerate data-driven research by enabling fast data experimentation and analysis (text and multimodal), and maintaining provenance of experimental results.
• Position results in existing infrastructure for human data collection, data analysis, data set creation and data visualization for exploration purposes. 
• Collaborate with other internal teams to ensure that Google DeepMind AI systems and products (e.g. Gemini) are informed by and adhere to the most advanced safety research and protocols
• Help make sure our AI models work well for everyone.

About you:

In order to set you up for success as a Research Scientist as part of DEER at Google DeepMind,  we look for the following skills and experience:

• You have at least one year of hands-on experience working with deep learning and/or foundation models (text and multimodal)
• You are adept at working with codebases that support machine learning at scale. You are familiar with ML / scientific libraries (e.g. JAX, TensorFlow, PyTorch, Numpy, Pandas), distributed computation, and large scale system design.
• Your knowledge of statistics, data analytics and machine learning concepts enables you to understand research papers in the field.
• You are excited to work with strong contributors from different fields of research to make progress towards a shared ambitious goal.
• You have experience in and enjoy working as part of interdisciplinary teams.

In addition, the following would be an advantage: 

• You have technical experience in responsible AI, sociotechnical AI and/or AI Safety (whether from industry, academia, coursework, or personal projects).
• You have experience in running evaluations and/or collecting human data.
• You have experience working with and handling datasets and designing creative dataset visualisations
• You have an interest in Natural Language Processing, image, video processing and related areas.
• You have experience in contributing to publishing research papers at conferences, ranging from ACL, NeurIPS, ICML, ICLR, to ACM CHI, or FAccT. 
• You are an open minded individual who has the ability to consider and actively listen to diverse perspectives, alternative ideas, and new approaches.

The US base salary range for this full-time position is $136,000 - $210,000 bonus benefits. Your recruiter can share more about the specific salary range for your targeted location during the hiring process.

Application deadline: 12pm GMT February 7th 2025 

Note: In the event your application is successful and an offer of employment is made to you, any offer of employment will be conditional on the results of a background check, performed by a third party acting on our behalf. For more information on how we handle your data, please see our Applicant and Candidate Privacy Policy.