Principal Machine Learning Engineer (Memory & Context)
Core
Designing and owning the architecture of agent memory and context engineering for Zillow's AI home-search assistant to enable long-term user understanding across multi-session conversations.
Role type
Principal Machine Learning Engineer (Agentic AI)
Builds
Production-ready agentic AI systems with persistent memory and optimized context handling for real estate search.
Domain
Real Estate / Agentic AI / LLM Systems
Deliverable
production ML models
Required skills
Large-scale ML system deployment, LLM-based agent building, agent orchestration, prompt engineering, context engineering, evaluation framework design, Python programming, technical leadership
Preferred skills
Agent/user memory system design, long-dialogue system experience, AI assistant personalization, publications/patents in memory or long-context models
Technologies
LangGraph, Agents SDK, AutoGen
Responsibilities
Own agent memory architecture (write, retrieve, update, retire), lead context engineering for production agents, build and orchestrate agents end-to-end, define evaluation frameworks for memory and context quality, stay on cutting edge of agentic paradigms, lead multi-team initiatives and communicate technical direction
Seniority
Principal, hands-on technical leader with mentorship
