Senior Data Scientist Third-Party Data Analytics & Evaluation

About Blend360

Blend is a premier AI services provider that combines data science, analytics, technology, and talent to help organizations solve complex business challenges. We partner with leading global companies to deliver data-driven solutions that drive measurable business impact.

Role Overview

We are looking for a Data Scientist with strong analytical skills and experience working with third-party data, data brokers, audience data, and data quality assessment, preferably within the AdTech or MarTech ecosystem.

In this role, you will evaluate external data providers, independently validate their data quality and performance claims, and transform validated datasets into standardized, analytics-ready features for household intelligence and downstream machine learning applications.

The ideal candidate is highly analytical, comfortable working with large and complex datasets, and proficient in Python and SQL for data analysis and validation. Experience with audience measurement, identity resolution, consumer data, advertising data, or marketing data platforms will be highly valuable.

Key Responsibilities

Third-Party Data Evaluation & Quality Assessment

• Evaluate third-party data providers, data brokers, audience datasets, panel data, and video/viewership data using real-world samples.
• Independently validate vendor-reported match rates, data coverage, label accuracy, and other data quality metrics.
• Analyze data completeness, duplication, bot activity, overlap, geographic distribution, and consistency across datasets.
• Develop and execute repeatable methodologies to assess vendor data quality and identify gaps between claimed and observed performance.
• Document findings, supporting evidence, and recommendations using standardized data quality scorecards.

Data Analytics & Statistical Validation

• Perform exploratory data analysis and statistical assessments to understand dataset characteristics, reliability, and usability.
• Analyze audience and household-level data distributions, including geographic segmentation such as Designated Market Areas (DMAs).
• Assess data overlap, sample representativeness, anomalies, and potential sources of bias.
• Define validation rules, quality thresholds, and measurable criteria for accepting or rejecting third-party datasets.
• Translate analytical findings into clear, actionable insights for technical and business stakeholders.

Data Mapping & Feature Engineering

• Map vendor-specific schemas, data dictionaries, and attributes to a common data taxonomy.
• Standardize and transform validated data into structured, reusable features for household intelligence and audience analytics.
• Contribute feature definitions, validation rules, and quality checks for a shared feature store supporting household-level analytics and machine learning.
• Collaborate with Data Engineers to define validated schemas, transformation requirements, and data pipeline specifications.
• Ensure that data is documented, traceable, reproducible, and suitable for downstream analytical and ML applications.

Identity Resolution, Privacy & Compliance

• Evaluate data matching methodologies, including deterministic and probabilistic identity resolution techniques.
• Work with identity-related concepts such as hashed PII, mobile advertising IDs (MAIDs), hashed emails (HEMs), household IDs, and audience identifiers.
• Assess identity match quality, duplication, household association, and potential inconsistencies across data sources.
• Identify potential privacy, consent, data provenance, and permitted-use concerns in third-party datasets, escalating issues to the relevant legal or compliance teams.
• Maintain clear documentation of data sources, assumptions, validation methods, and limitations.

Stakeholder Collaboration & Reporting

• Partner with Data Engineers, Data Scientists, and business stakeholders to operationalize validated datasets.
• Present quantitative findings and vendor assessments in a clear, concise, and business-friendly manner.
• Provide evidence-based recommendations on vendor data suitability, quality risks, and potential use cases.
• Support continuous improvement of data evaluation methodologies, quality standards, and analytical processes.

Required Experience & Skills

• 4–8 years of experience in data science, data analytics, data quality analysis, or a related quantitative role.
• Strong experience analyzing third-party datasets, data broker data, audience data, consumer data, panel data, or other externally sourced datasets.
• Exposure to the AdTech or MarTech ecosystem, including audience data, advertising measurement, marketing analytics, identity resolution, or customer data platforms.
• Strong analytical and statistical skills, with experience in data profiling, match-rate analysis, deduplication, overlap analysis, and distributional assessment.
• Proficiency in Python for data analysis and manipulation, particularly using pandas and NumPy; scikit-learn or equivalent libraries are beneficial.
• Strong SQL skills for data extraction, joining, aggregation, reconciliation, and validation across large datasets.
• Experience evaluating data dictionaries, schemas, source provenance, completeness, consistency, and reliability.
• Understanding of deterministic versus probabilistic matching and familiarity with hashed PII, MAIDs, HEMs, and household-level identifiers.
• Working knowledge of feature engineering and preparing structured datasets for downstream analytics or machine learning.
• Strong documentation and communication skills, with the ability to explain quantitative findings to non-technical stakeholders.

Preferred Qualifications

• Previous experience working with data brokers, third-party data vendors, audience intelligence platforms, or consumer data providers.
• Familiarity with AdTech and MarTech platforms, advertising measurement, audience segmentation, identity graphs, or household-level intelligence.
• Exposure to data onboarding, identity matching, audience enrichment, campaign measurement, or marketing attribution datasets.
• Experience evaluating video viewership, panel, publisher, device, or household-level datasets.
• Familiarity with data privacy, consent management, data governance, and responsible use of third-party data.
• Experience working with cloud-based data platforms, data warehouses, or feature stores.
• Exposure to machine learning, embeddings, or household-level feature representations.