Zillow is seeking to develop advanced AI-driven customer experiences that leverage large language models (LLMs) and multi-agent systems to transform the way people buy, sell, rent, and finance homes.
Requirements
- Expertise in agentic AI, including pretraining, fine-tuning, and reinforcement learning of large language models.
- Strong programming skills in languages such as Python, and familiarity with AI frameworks and cloud infrastructure.
- Experience working with multi-agent systems, collaboration frameworks, or similar paradigms.
- Proven experience with deploying and scaling AI services capable of handling hundreds of millions of interactions daily, with high availability and low latency.
- Familiarity with AI frameworks and cloud infrastructure.
- Experience with data pipelines and dataset management.
- Knowledge of evaluation frameworks and metrics for model performance.
Responsibilities
- Build and maintain data pipelines for training and evaluating large language models, ensuring data quality, privacy, and proper dataset management.
- Curate user understanding signals, such as intents, preferences, and behavioral features, to inform model development.
- Develop and manage labeling processes, including heuristics, annotation jobs, and prompt-based labeling, to generate high-quality training data.
- Design, prototype, and deploy agentic AI solutions, including multi-agent systems utilizing frameworks like LangGraph.
- Implement context-aware features in collaboration with senior engineers to enhance AI agent responsiveness and personalization.
- Establish evaluation frameworks to measure model performance on accuracy, bias, safety, and user intent coverage, and develop dashboards for ongoing monitoring.
- Lead experimentation efforts by designing metrics, conducting A/B tests, and analyzing results to improve model reliability and effectiveness.
Other
- A master’s degree or higher in Computer Science, Electrical Engineering, or a related field, or equivalent practical experience.
- Strong problem-solving skills, ability to work independently and collaboratively, and excellent communication skills.
- Ability to publish high-impact research or articles in top AI conferences is a plus.
- Collaborate with cross-functional teams, participate in code reviews, and share best practices to foster a culture of continuous improvement.
- Summarize research findings and model evaluations into clear documentation and presentations for internal and external stakeholders.
