Sr Director of AI Search & Recommendations

Search is the front door to Realtor.com.

Your mandate: make our AI search & discovery experience the most trusted, most relevant, and fastest path to the right home—and do it at public-company scale. As VP of Engineering, AI Search & Discovery, you’ll lead a multidisciplinary organization across AI Search Engineering, Relevance/Ranking, Query Understanding, Personalization, and ML Platforms, partnering tightly with Product, Data, and Revenue to move the company’s most important customer and business metrics.

This is a hands-on Sr Director role: you’ll define the strategy, shape the architecture, hire and develop the leaders, and hold a very high bar on execution, reliability, and measurable outcomes.

What You'll Do

Set the bar and the direction

• Own the 3 year technical strategy for AI search, discovery, and recommendations; align it to company OKRs and News Corp governance.
• Establish a metrics-first culture: nDCG/ERR, CTR, lead quality, P50/P95 latency, experiment velocity, and infra cost per search.

Build & lead a world-class engineering organization

• Lead the team across core search, ranking, query understanding, personalization, and ML platform; grow and coach an org spanning software engineers, relevance scientists, and ML engineers.
• Raise the talent bar: recruit senior technical leaders and principal-level ICs; set clear career paths and performance standards.

Own the search stack end-to-end

• Retrieval & indexing: hybrid vector + semantic, data quality, dedupe, freshness, and geo/local considerations.
• Ranking: multi-objective LTR (relevance, quality, monetization), session & user modeling, cold-start handling for new listings, image and text quality signals, and advanced re-ranking (including LLM-assisted ranking where it wins).
• Query understanding: intents, entities, spell-correction, reformulation, facets, structured constraints, ambiguity resolution for neighborhoods/communities.
• Personalization: embeddings & features, eligibility/guardrails, unbiased evaluation, and robust online/offline experimentation pipelines.
• Evaluation & experimentation: offline suites, interleaving, online A/B, sequential testing; ship improvements continuously with safety rails.

Modernize the platform

• Shape our ML & inference platform for low-latency ranking and re-ranking
• Partner with platform teams on SLOs, cost, capacity planning, observability, and incident response (you’ll expect clean postmortems and durable fixes).
• Advance pragmatic GenAI: RAG for search assistance, answer synthesis with citations, robust eval harnesses (task-specific metrics, human-in-the-loop), and cost/perf optimization.

Operate with rigor at a public company

• Work with Legal/Privacy/Compliance on MLS data obligations, Fair Housing considerations, and AI governance; ensure robots/agents compliance and abuse/spam defenses.
• Communicate with executive leadership and, when needed, Boards/partners; convert complex trade-offs into crisp, defensible decisions.

You will champion the adoption of AI-native workflows, driving team efficiency through the strategic implementation of coding assistants and automated debugging tools. This role requires fostering a culture of AI literacy, where you will mentor the team on effective prompting strategies and the ethical integration of LLMs into the SDLC. You are expected to establish guardrails for code security and quality assurance within AI-assisted workflows and continuously evaluate emerging tooling to keep our engineering practice at the cutting edge

• Proven scale: You’ve led search/recommendation systems serving tens of millions of users or equivalent high-stakes marketplaces, with measurable lifts in relevance and revenue.
• Depth in search & ML: 10+ years in large-scale search/recommendation, including hybrid retrieval, learning-to-rank, user modeling, and evaluation. You’ve shipped multi-objective ranking safely in production.
• Bachelor’s degree in Computer Science, or equivalent experience
• GenAI pragmatism: Hands-on leadership integrating LLMs where they improve outcomes (e.g., re-ranking, query reformulation, answer synthesis) with strong offline/online evaluation discipline.
• Platform and operations chops: Real-time inference, feature stores, streaming data, strong SRE partnership, and a habit of designing for latency, reliability, and cost.
• People & org leadership: Built and scaled multi-disciplinary teams; hired/developed Directors and Principals; set culture, strategy, and execution cadence.
• Executive communication: Clear storyteller able to align product, design, science, finance, legal, and revenue stakeholders on trade-offs and timelines.
• Austin presence: Willing and able to work onsite at our Austin HQ (relocation required to make Austin your primary residence). 

• Core Search (indexing, retrieval, query understanding)
• Ranking & Relevance Science (features, models, re-ranking, guardrails)
• Personalization (session/user modeling, eligibility, privacy-aware features)
• Evaluation & Experimentation (offline/online, human-in-the-loop)
• ML/Inference Platform for search use-cases (latency-critical APIs, feature store)
• Quality & Integrity (spam/fraud/abuse defenses; content quality signals)

• Quality: Meaningful gains in nDCG/ERR and conversion/lead quality; improved first-result time and P95 latency on key surfaces.
• Velocity: 2–3× increase in experiment throughput; reliable weekly relevance drops with automated evaluation.
• Platform: Real-time re-ranking and feature pipeline in place; safe & measurable GenAI integrations where they outperform baselines.
• Engineering Organization: Fully staffed leadership bench; clear succession and hiring plan; healthy engagement and retention.
• Operational excellence: SLOs met, sharp incident posture, and a cost curve that bends the right way.