The Mosaic AI organization enables companies to develop AI models and systems using their own data, with technologies ranging from fine-tuning LLMs for enterprise domains, to a platform for building compound AI systems that use retrieval and agents. Mosaic AI is committed to the belief that a company's AI models are just as valuable as any other core IP, and that high-quality AI models should be available to all.

Job description:

Most of the world's data+AI problems lie in enterprise domains, behind closed doors. Our research team's goal is to push the frontier of 'domain adaptation' - how can we develop LLMs and AI systems that work well for custom domains. To do this we are tackling open research problems on a range of topics, from how to scale/automate eval, fine tune with synthetic data, retrieval augmentation, fast/efficient inference and more.

You will work with our research team on projects focused on adapting LLMs and AI systems towards enterprise domains. This may include:

• Adapting, improving, and evaluating a method from the literature.
• Designing an entirely new method for domain adaptation.
• Composing together multiple methods to create new recipes for efficient post-training.
• Evaluation of LLMs and AI systems.

Your qualifications and qualities:

• Required:
• Research experience in and proficiency with the fundamentals of deep learning.
• Pursuing a PhD in computer science or related fields (electrical engineering, neuroscience, physics, math, etc.).
• Proficient software engineering skills, including with PyTorch.

Pay Range Transparency

Databricks is committed to fair and equitable compensation practices. The pay range(s) for this role is listed below and represents the expected salary range for non-commissionable roles or on-target earnings for commissionable roles. Actual compensation packages are based on several factors that are unique to each candidate, including but not limited to job-related skills, depth of experience, relevant certifications and training, and specific work location. Based on the factors above, Databricks anticipates utilizing the full width of the range. The total compensation package for this position may also include eligibility for annual performance bonus, equity, and the benefits listed above.

XML job scraping automation by YubHub

At DeepMind, we're a team of scientists, engineers, machine learning experts and more, working together to advance the state of the art in artificial intelligence. We use our technologies for widespread public benefit and scientific discovery, and collaborate with others on critical challenges, ensuring safety and ethics are the highest priority.

This is a high impact role that will impact the efficiency of serving through very low precision and sparse models. Additionally, this role will help drive the future Google HW roadmap to support forward looking numerics. This role offers the unique opportunity to address a historically underserved but increasingly critical area in the advancement of AI.

Responsibilities:

• Decide the precision, numerics, and sparsity formats used by key Google AI models and ensure these decisions are reflected in the roadmaps for corresponding hardware.
• Drive novel research advancements and bring the most impactful ideas into production.

About You:

To set you up for success as a Software Engineer at Google DeepMind, we look for the following skills and experience:

• PhD in Computer Science or related field with at least 2+ years of relevant experience.
• Strong software-engineering skills in addition to a research background.
• Deep understanding of the numerics/quantization/sparsity literature.
• Practical experience driving low precision and sparse models through to production.

XML job scraping automation by YubHub