Principal Software Engineer, Applied AI Services

About the role

Shopper Journey Services (SJS) is Zillow's engine for authenticated user intent and a core enabler of Home Details Page (HDP) experiences. We gather rich, explicit and implicit signals from authenticated shoppers and co‑shoppers, refine them into durable models, services, and display‑ready data, and activate those capabilities across Zillow's ecosystem to deliver personalized, high‑quality experiences. Our charter spans two tightly connected areas: User Intent & Applied AI, where we build the pipelines, services, and evaluation capabilities that turn user signals into intelligence and AI‑powered experiences, and the HDP Backend Platform, where we power HDP with robust, low‑latency, display‑ready data. AI is both a product capability and an engineering accelerator for this team, and we are investing in LLM‑ and ML‑powered workflows, evaluation, and agentic systems while modernizing our conventional backend stack. We are looking for a P5 Principal Software Engineer, Applied AI Services to lead the architecture and evolution of our backend and AI systems across Shopper Journey Services and partner teams. In this cross‑cutting role, you will operate at the intersection of large‑scale backend systems (microservices, APIs, data stores, event‑driven architectures, ZGCP/Kubernetes, AWS‑class cloud infrastructure) and applied AI/ML systems (offline data ingestion, feature and signal pipelines, LLM/ML‑powered capabilities, evaluation frameworks, and AI‑driven workflows). You will play a leading role in how we bring AI capabilities into production services safely and repeatably, evolve our user intent flywheel and HDP backend platform, and define system‑wide standards, frameworks, and abstractions that enable many teams to ship AI‑powered features faster and with higher quality. This is a hands‑on principal IC role where you will both design and build systems yourself and provide technical leadership across multiple teams and orgs, without direct reporting lines. You Will Get To Architect end‑to‑end applied AI services that connect offline data ingestion, AI/ML/LLM workflows, and online services and APIs, defining shared patterns for batch and streaming data pipelines (e.g., Databricks, Spark, Kafka or equivalents), feature and signal stores, and evaluation and guardrail frameworks for AI‑powered capabilities. Create reusable building blocks-such as libraries, templates, and reference implementations-that make it straightforward for product teams to integrate AI into their services and ship AI‑powered features faster. Lead conventional backend and platform excellence by architecting and guiding high‑scale microservices in a Kubernetes environment, driving patterns for event‑driven architectures (including event schemas, contracts, and consumption patterns), and setting standards for databases, caching, and data access that support both HDP backend and AI use cases. Drive AI/ML and LLM‑powered systems from prototype to production, including ingestion and transformation of training and inference data, integration of models and LLMs into online decision flows and APIs, and the definition of evaluation methodologies, metrics, and regression gates (e.g., LLM‑as‑judge, offline/online evaluation, human‑in‑the‑loop review loops). Partner with AI/ML, Agentic AI, and data platform teams to clarify ownership boundaries and interfaces (for example, around cross‑cutting evaluation capabilities such as Evaluate MCP), and to ensure AI systems remain measurable, debuggable, and reproducible as they scale. Lead multi‑team technical initiatives that span SJS, AI/ML teams, HDP, and other backend groups, defining and rolling out system‑wide standards and abstractions for APIs and contracts (REST/GraphQL, events, DRDCs), data schemas and lineage across offline and online paths, and observability, evaluation, and operational runbooks for AI‑powered services. Mentor senior engineers, run deep design reviews, and champion the use of AI as a force multiplier for engineering-such as background agents for KTLO (library updates, security posture, config drift) and AI‑assisted design, implementation, and testing patterns-helping define which workflows should be agent‑assisted versus human‑led in safe, observable, and cost‑effective ways. This role has been categorized as a teleworker position. Teleworkers do not have a permanent corporate office workplace and instead work from a physical location of their choice, which must be identified to the Company. Employees may live anywhere in Mexico, with availability to travel to Mexico City, as we recommend attendance at occasional office events. In addition to a competitive base salary and benefits, this position is also eligible for equity awards based on factors such as experience, performance and location.

Requirements

• You have 10+ years of software engineering experience with a strong track record of delivering and scaling complex, distributed backend systems

Benefits

Your Match

How well this role fits your profile.

Company Intel

What employees say

Worked at Zillow? Share your experience