Senior Data Scientist II, Ranking & Relevance

Fetch•US - Remote,

1d•$183,665 - $216,077

About The Position

The Senior Data Scientist II will own Fetch’s analytical and experimentation roadmap for ranking and relevance, partnering with Product, Engineering, and ML to improve advertiser outcomes, user value, and Fetch’s long-term economics. As the team’s dedicated data scientist, this person will set the data science strategy for relevance and ranking and work with engineers on models that drive user engagement, rewards earned, and partner economic outcomes. The role develops frameworks that balance advertiser incrementality and delivery, user relevance and retention, and platform revenue, reward costs, and margin. The role is the core data science partner for go-to-market, offer development, and offer delivery teams. It translates commercial goals into measurable objectives, product requirements, and ranking improvements; informs senior stakeholders’ launch, iteration, or stopping decisions; and mentors others while establishing reusable methods that improve scientific quality and execution. Fetch’s AI & Data organization works across data engineering, analytics engineering, machine learning, experimentation, and data platforms to turn data into business and customer impact. Its practitioners contribute hands-on to production systems, analytical foundations, and intelligent features, collaborate with product, platform, and engineering partners, and help establish standards and practices that support reliable scale.

Requirements

At least 8 years of experience in data science, machine learning, or applied analytics, with demonstrated ownership of complex analytical problems in product-driven environments.
Experience owning the analytical roadmap and experiments for ranking, relevance, or recommendations.
Ability to explain which questions were prioritized, which work was declined, and why.
Cross-functional collaboration experience involving product, sales, and engineering.
Experience defining marketplace KPI frameworks and resolving advertiser–user–platform trade-offs.
Track record of evaluating how improvements to individual metrics affect overall marketplace economics.
Experience designing measurement that accounts for selection bias, shared budgets, or spillovers.
Ability to defend experiment design and assumptions, addressing selection bias, interference, or observational identification.
Demonstrated incremental, persistent business impact.
Track record of influencing strategy or launch decisions.
Examples of redirecting, pausing, or stopping initiatives based on quantified trade-offs.
Experience building reusable analytical assets and mentoring analysts or scientists.
Strong modeling expertise.
Strong proficiency in SQL and at least one programming language, preferably Python.
Experience with large-scale data systems such as Snowflake, dbt, Airflow, or Spark.

Nice To Haves

Substantial experience with advertiser-funded consumer marketplaces, retail media, or ad platforms balancing advertiser, user, and platform incentives.
Experience in ad ranking or product search and ranking.
Consumer product analytics experience connecting behavior to retention and monetization.

Responsibilities

Own the data science strategy and roadmap for ranking and relevance as the team’s dedicated data scientist.
Partner with engineers to improve ranking models end to end, from defining optimization objectives through experimentation.
Own experiment design and convert results into clear launch recommendations.
Design marketplace measurement that accounts for bias and interference between advertisers.
Use rigorous observational methods when experiments are not possible and defend the assumptions behind them.
Build marketplace frameworks that balance advertiser, user, and platform value, with clear success criteria and trade-off guardrails.
Distinguish incremental gains from value shifted between advertisers, surfaces, or time periods.
Measure the effects of changes across user and advertiser segments, distinguishing durable value from short-term lift.
Use findings to shape targeting and investment strategy while accounting for uncertainty.
Quantify commercial impact to guide senior leaders’ priorities and product decisions.
Partner with commercial and offer teams to translate business goals into measurable objectives and ranking improvements.
Communicate with technical and non-technical partners.
Mentor others and create reusable methods that improve scientific quality and execution.
Use tools and technologies such as Python, SQL, Snowflake, dbt, Airflow, Spark, and AWS to conduct and scale analytical work.