About the job
As a PhD Research Intern on the AQ/EQ Foundation team, you will conduct state-of-the-art research in foundational models and building agentic AI systems. You will focus on fine-tuning and reinforcement learning of large language models (LLMs), with an emphasis on customizing them for Zillow’s domain. You’ll also explore the design of automated, agentic workflows that allow intelligent systems to reason, plan, and act in ways that directly improve the home-buying experience.
Responsibilities
Researching and developing techniques for fine-tuning LLMs with domain-specific data
Applying reinforcement learning to optimize model performance for user-centric outcome
Designing and prototyping agentic workflows that can autonomously perform tasks and assist home buyers
Collaborating with cross-functional teams to evaluate and deploy research prototypes
Sharing insights through presentations, documentation, and potentially publications
Qualifications
Minimum
Currently enrolled in a PhD program in Computer Science, Machine Learning, Artificial Intelligence, or a related field with a strong publication record
Advanced research in natural language processing (NLP) and/or reinforcement learning (RL)
Practical experience fine-tuning and adapting large language models (LLMs) for specific use cases
Familiarity with the design and implementation of automated/ agentic workflows
Deep understanding of LLMs, hands on experience of post-training with the most popular OSS models
Proficiency in Python and modern ML frameworks (e.g., PyTorch, TensorFlow)
Preferred
Excited about applying advanced AI methods to impactful, real-world problems
Strong communication skills and ability to work collaboratively in a multidisciplinary environment
Strong research mindset, with motivation to publish