Data Scientist
Posted 9 Oct 2026
Advertised as “Data Scientist, Mobile App Performance ”
BranchFactor Summary
The original posting with the fluff stripped out
- Expected experience
- 5+ yrs
- Management level
- Individual contributor
- Employment
- Hybrid
No salary listed for this role.
Field of studyComputer Science, Mathematics, Statistics (or related)
Study levelMaster's or higher
About the Role:
Own the data science function for Mobile App performance, working at the intersection of ML and programmatic advertising. This is a high-ownership, high-visibility role--both internally and with our customers.
What You'll Do:
Model Development
-
Build signal enrichment pipelines leveraging contextual signals, behavioral data, and mobile attribution frameworks
-
Develop production-grade predictive models based on event data (clicks, installs, in-app events, retention).
-
Partner with GTM teams to translate metric frameworks into buyer-facing insights and predictive models
Drive Product Development
-
Conduct deep-dive analysis to influence product and engineering decisions: pull and query large datasets, extract valuable signals from noise, and produce actionable insights into how we can drive better outcomes for buyers and publishers
-
Develop predictive modeling frameworks to quantify performance gains from product hypotheses before they ship — reducing decision latency and de-risking the roadmap
-
Act as the senior data science voice in product discussions. Bring partner-level insight and analytical rigor to roadmap decisions.
Evaluate and Improve System Health
-
Understand changes and drivers across PubMatic's mobile ecosystem: diagnose issues, identify structural inefficiencies, and track input/output metrics at each system component
-
Work with product and engineering to define success metrics, design experiments, and evaluate new features before and after launch.
-
Design experiment frameworks and causal inference methods that produce reliable, unbiased signals for product and business decisions
Who You Are:
Required
-
5+ years of hands-on data science experience, with a focus on ML model development, statistical inference, and production-grade analysis
-
Deep fluency in Python and SQL; experience with large-scale data environments (Spark, Hadoop, Hive, or equivalent)
-
Strong foundation in experimental design, hypothesis testing, regression modeling, and causal inference
-
Experience with ML algorithms relevant to ranking, prediction, and optimization problems: gradient boosted trees, deep learning, k-means, and related methods