Product Data Scientist - SFO, CA - Hybrid

We need to output HTML fragment with sections: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule only if content present. The posting includes: About The Role (prose), What You Ll Do (list of duties), About You (experience, education, skills), Nice to Have (list), Time Zone Requirements (maybe schedule? but not pay/benefits). No explicit Benefits, Pay, Schedule. So we should include sections: About the role (maybe combine About The Role and About You? Actually About The Role is a section; Responsibilities is What You Ll Do; Requirements/Qualifications/Skills could be derived from About You and Nice to Have. We need to decide mapping: The instruction: use headings for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Skills). So we need to map content accordingly. We have About The Role prose -> heading

About the role

with

. Responsibilities: list of duties from "What You Ll Do". Use

• . Requirements: maybe core requirements from "About You" (experience, education, etc). Could be a list. Qualifications: maybe similar to Requirements? Might be redundant. But we can split: Requirements (must-have) and Qualifications (nice-to-have) or Skills. However instruction says prefer few substantive sections over many thin ones. We could combine Requirements and Qualifications into one? But they ask to use headings if posting has real specific content. We have both mandatory and nice-to-have. Could make sections: Requirements (mandatory) and Qualifications (nice-to-have). Skills maybe also covered. But to avoid many thin sections, we could have Requirements (including both mandatory and nice-to-have) as a list. But they specifically list possible headings; we can choose which to include. We have no Benefits, Pay, Schedule explicit. So omit those. Thus output:

  About the role

  ... (the About The Role paragraph)

  Responsibilities

  • ... each bullet from What You Ll Do

  Requirements

  • ... mandatory requirements from About You

  Qualifications

  • ... nice-to-have items
  . Ensure we keep punctuation. Now Requirements from "About You": Let's parse: "Experience in a hybrid analytics/data science role (e.g., analytics consulting, product data science, applied statistics) with a track record of directly informing product decisions; bachelor s degree or higher in a quantitative field, or equivalent combination of education and experience. You have strong grounding in statistics and experimentation hypothesis testing, causal inference, experiment design, and you can explain the difference between a significant result and a meaningful one. You re fluent in SQL and at least one scripting/statistical language (Python or R), and you re comfortable enough with data engineering fundamentals (pipelines, transformations, data modeling) to build what you need and partner effectively with engineers on the rest. You can develop, validate, and communicate the tradeoffs of statistical and machine learning models, and you know when a simpler model or a well-designed experiment beats a complex one. You use AI tools in your day-to-day work for exploratory analysis, documentation, and accelerating routine analytics and you know when their outputs need scrutiny before they touch a product decision. You think like a consultant: you get to the real question behind the question, structure ambiguous problems, and land on recommendations stakeholders can act on. You have good judgment about rigor versus speed, and you don t cut corners on measurement integrity just to hit a deadline. You re a clear communicator who can flex between a technical conversation with engineering and a decision-focused conversation with product and business stakeholders. You re curious about how data, experimentation, and AI can change what s possible in consumer lending products, and you re always looking for a better way to answer the question." We need each as
  • . Keep as is. Now Qualifications (Nice to Have): list: "Background in fintech, consumer lending, or home improvement/contractor financing. Experience with CDP platforms, event instrumentation tooling (e.g., Segment, mParticle, Amplitude), or experimentation platforms. Hands-on experience with credit or risk modeling, pricing strategy, or marketing decisioning. Experience with dbt, Airflow, or similar data pipeline/orchestration tools. Prior experience embedded directly with product teams in an agile/scrum environment." Also maybe "Time Zone Requirements Flexible, with core overlap expected with Pacific Time hours" could be considered Schedule? But schedule not given; it's about timezone. Could be considered Schedule? The instruction: include Schedule section only if posting actually states them. It does state timezone requirement. Could treat as Schedule? But they said Schedule section for schedule info like hours, shift. Time zone requirement maybe akin to schedule. However they said "never create Pay/Schedule/Benefits sections unless the posting actually states them." It does state timezone requirement, which is about schedule overlap. Could be considered Schedule. But we need to decide. The posting includes "Time Zone Requirements Flexible, with core overlap expected with Pacific Time hours". That's a schedule-related detail. So we could include a

    Schedule

    with

    Time Zone Requirements Flexible, with core overlap expected with Pacific Time hours.

    . That seems appropriate. We should not include Benefits or Pay. Now ensure we don't repeat job title as heading. Not include Job ID etc. Now produce HTML fragment. We need to escape any HTML special characters? The text includes apostrophes, etc. It's fine as plain text. We'll output:

    About the role

    This role sits at the intersection of product analytics, experimentation, and data science embedded directly with Product Management to help shape and grow our Home Improvement lending product. You ll be the analytical backbone for a product team making high-stakes decisions about underwriting funnels, borrower experience, and growth, translating messy data into clear evidence and, where it counts, into models and instrumentation that ship. It s a high-impact role for someone who is equally comfortable running a rigorous A/B test, writing a dbt model, and explaining a lift curve to a VP.

    Responsibilities

    • Partner day-to-day with Home Improvement Product Managers as their embedded data science and analytics resource turning open-ended product questions into structured analyses and clear recommendations.
    • Design, run, and interpret experiments (A/B and quasi-experimental) across the borrower funnel from offer presentment through origination with rigor around power, sample ratio mismatch, novelty effects, and interaction risk across concurrent tests.
    • Define and own the metrics framework for the Home Improvement product line: north-star and guardrail metrics, funnel and cohort definitions, and the instrumentation needed to measure them reliably.
    • Work with engineering to ensure event tracking and logging are complete, accurate, and well-documented at the point of instrumentation, not discovered as gaps after the fact.
    • Build and maintain data pipelines and models (e.g., SQL/dbt transformations, feature pipelines) well enough to be self-sufficient for most analyses and to collaborate credibly with data engineering on the rest.
    • Develop and validate statistical and ML models supporting product decisions response/propensity models, funnel drop-off and conversion models, segmentation, and early-stage risk or pricing signals in partnership with credit strategy with attention to fairness, explainability, and regulatory context appropriate to a lending business.
    • Bring AI fluency to the work: use LLM- and agentic-tooling to accelerate exploratory analysis, requirements gathering, and documentation, while knowing where automated outputs need human judgment and validation before they inform a decision.
    • Communicate findings in a way that drives action clear write-ups, well-chosen visualizations, and recommendations tied to specific product or roadmap decisions, not just descriptive dashboards.
    • Contribute to PI planning and roadmap discussions by sizing opportunities, flagging measurement risk in proposed initiatives, and helping the team commit to work that can actually be evaluated.
    • Continuously monitor product and experiment performance post-launch, and proactively surface anomalies, regressions, or new opportunities rather than waiting to be asked.

    Requirements

    • Experience in a hybrid analytics/data science role (e.g., analytics consulting, product data science, applied statistics) with a track record of directly informing product decisions; bachelor s degree or higher in a quantitative field, or equivalent combination of education and experience.
    • You have strong grounding in statistics and experimentation hypothesis testing, causal inference, experiment design, and you can explain the difference between a significant result and a meaningful one.
    • You re fluent in SQL and at least one scripting/statistical language (Python or R), and you re comfortable enough with data engineering fundamentals (pipelines, transformations, data modeling) to build what you need and partner effectively with engineers on the rest.
    • You can develop, validate, and communicate the tradeoffs of statistical and machine learning models, and you know when a simpler model or a well-designed