Marketing Science Product Engineer

We are a tech-enabled growth firm–at the intersection of marketing, consulting & data intelligence–igniting revenue and brand recognition for leading and emerging companies around the world. As a people-first firm, we value diversity in backgrounds and experiences. As a full-service growth marketing firm, we offer best-in-class services including: SEO, Content Marketing, Paid Media, Social Media Marketing, Programmatic + CTV, Public Relations, Influencer Marketing, Email + SMS, Conversion Rate Optimization, Retail Marketing, and Creative. Here at Power Digital, we are hyper-focused on helping brands drive revenue growth and brand recognition, ultimately driving irrefutable value for our clients. At the heart of Power Digital is our proprietary technology, nova, which analyzes businesses through first-party data, simplifying investment planning for marketing and diligence in M&A—putting marketers in a strategic seat at the table—and providing value in unparalleled ways. Managing billions in media, our dynamic team––of consultative marketers, creatives, analysts and technologists––challenge traditional ways of planning and measurement through meticulous testing and data science across each milestone of the customer journey. Proficiency in spoken and written English at an advanced level is required for this role. A day in the life: As a Marketing Science Product Engineer, you’ll work at the intersection of data science and engineering, transforming our MMM & causal frameworks into production-ready Python utilities used across teams and our UI environments. You will partner closely with the Director of Marketing Science - Product, marketing scientists, and engineering collaborators to turn statistical specifications into scalable, maintainable code. This is not an analytics role; it is a product builder role for causal measurement in Python. This role executes against Product-defined specifications. All decisions related to product scope, methodology changes, metrics definitions, acceptance criteria, and rollout timing remain with Product Leadership. The role is expected to surface risks, ambiguities, or improvement opportunities, but not to independently redefine specifications. Responsibilities: Implement MMM modules such as: Sequential Cross-Validation (SCV) utilities, Halo/cannibalization & revenue allocation logic, Parameter QC to detect overfitting, shrinkage misuse, leakage, etc. Scenario Planning engine, Science and Engineering for out-of-the-box attribution based on MMM variations. Convert notebook based workflows into reusable and tested Python modules. Build functions to automate Meridian pipelines. Causal Measurement & Feature Development: Build statistical utilities for: Diminishing-return and saturation curves, Efficiency quadrants & spend allocation plots, Posterior diagnostic visualizations and stability checks. Implement and support product-approved attribution methodologies. Workflow Automation & Registry: Develop, automate and maintain: Metadata schemas for MMM model registry, Versioning utilities and standardized logs, Data audit & EDA standardization modules. Integrate registry logging with Snowflake or other data stores. Work with leadership to translate specifications into code. Maintain clear documentation, examples, tests and onboarding material. Support the enablement of measurement analysts using standardized utilities. Bridge Between Methodology, Product, and Modelers: Serve as a technical connector between Product and the Marketing Science modeling team, ensuring new features are usable, scalable and aligned with methodology standards. Help onboard modelers, enabling consistent usage of utilities, automated diagnostics, metadata logging and scenario tools. This role surfaces feedback and implementation risks but does not independently define or modify product logic, methodology, metrics or rollout decisions. Testing, QA & Reliability: Write and maintain unit tests, regression tests, and reproducibility checks for MMM packages and utilities. Validate code performance across multiple brands and modeling conditions (different datasets, geographies, SKU structures, sales cycles). Ensure SCV, diagnostics, metadata, and attribution modules are stable, performant, and free of methodological regressions. Document edge cases and error handling recommendations. Role Requirements: 5 years in data science, data engineering and machine learning engineering. Very strong proficiency in Python , including:Modular package design & clean function architecture Pandas, NumPy, SciPy, serialization, project organization Versioning, environment management, testing frameworks Experience building time-series and causal models . Familiarity with Media Attribution and Bayesian modeling . Experience building reusable plotting and BI utilities such as Looker Studio, Matplotlib, Plotly and Seaborn . Prior work with SQL, Snowflake or GCP, ideally with Google Cloud Run. Strong communication skills for explaining model mechanics and code behavior. Key Performance Indicators (KPIs) Methodology → Code KPI: % of methodology specifications delivered as reusable modules >90% of new specs implemented within agreed timeframes Reusable Automation KPI: Reuse rate of modules across client models Target: >90% of active MMM projects use the standardized utilities QA & Stability KPI: Test coverage + regression success rate Target: >80% code test coverage and 90% feature adoption among modelers +