Data Specialist
PALNAR, New York, NY, United States
7+ years of applied data science experience, with at least 4 years focused on MMM and marketing/media measurement
Production ownership: you have built and shipped data pipelines and model-serving integrations – not just notebooks handed off to engineering (be ready to talk through specific systems you have put into production and the business outcomes they drove)
Strong Python/R and SQL skills; experience using modern data stack (cloud data warehouses, orchestration frameworks, version control, CI/CD)
Demonstrated experience with incrementality testing, lift studies, and MTA, including the messy parts: test design, power analysis, reconciling conflicting signals across methodologies
Experience with at least one major MMM framework, such as Meridian, PyMC, Robyn, or LightweightMMM, and can speak to the tradeoffs in going with a bespoke model versus using such frameworks
Proven ability to own stakeholder communication end-to-end: presenting to senior internal leadership (VP+), agencies, and clients without needing a layer of translation (you move swiftly between a Bayesian prior conversation and a channel-mix recommendation conversation without skipping a beat)
Able to work hybrid from our NYC office (4 days/week).
Strongly Preferred
Prior experience at a publisher, broadcaster, streaming platform, ad-tech company, or agency/measurement firm working on media-side measurement
Familiarity with ACR data, set-top box data, or digital/streaming viewership data
Experience working with 3P conversion/outcome data providers (NIQ, Polk, EDO, or similar)
Exposure to both pre-sales (research, planning, recommendation) and post-sales (outcomes, renewal) measurement workflows
GCP experience (BigQuery, Vertex AI, Composer).
Nice-to-haves
Causal inference depth: geo-based measurement (GeoLift, CausalImpact), synthetic controls, uplift modeling
Experience with clean room environments (Snowflake, AWS, Habu, InfoSum)
Familiarity with identity resolution and cross-platform measurement
#J-18808-Ljbffr
Production ownership: you have built and shipped data pipelines and model-serving integrations – not just notebooks handed off to engineering (be ready to talk through specific systems you have put into production and the business outcomes they drove)
Strong Python/R and SQL skills; experience using modern data stack (cloud data warehouses, orchestration frameworks, version control, CI/CD)
Demonstrated experience with incrementality testing, lift studies, and MTA, including the messy parts: test design, power analysis, reconciling conflicting signals across methodologies
Experience with at least one major MMM framework, such as Meridian, PyMC, Robyn, or LightweightMMM, and can speak to the tradeoffs in going with a bespoke model versus using such frameworks
Proven ability to own stakeholder communication end-to-end: presenting to senior internal leadership (VP+), agencies, and clients without needing a layer of translation (you move swiftly between a Bayesian prior conversation and a channel-mix recommendation conversation without skipping a beat)
Able to work hybrid from our NYC office (4 days/week).
Strongly Preferred
Prior experience at a publisher, broadcaster, streaming platform, ad-tech company, or agency/measurement firm working on media-side measurement
Familiarity with ACR data, set-top box data, or digital/streaming viewership data
Experience working with 3P conversion/outcome data providers (NIQ, Polk, EDO, or similar)
Exposure to both pre-sales (research, planning, recommendation) and post-sales (outcomes, renewal) measurement workflows
GCP experience (BigQuery, Vertex AI, Composer).
Nice-to-haves
Causal inference depth: geo-based measurement (GeoLift, CausalImpact), synthetic controls, uplift modeling
Experience with clean room environments (Snowflake, AWS, Habu, InfoSum)
Familiarity with identity resolution and cross-platform measurement
#J-18808-Ljbffr