Senior Data Scientist, Identity Graph

Impact.com•Columbus, OH

About The Position

We're looking for a Senior Data Scientist with deep expertise in identity graph construction and resolution to lead a critical modernization of Impact's identity graph infrastructure. This is a technically demanding individual contributor role — you'll own the diagnostic, architectural, and implementation work required to understand our current identity graph end-to-end, identify its weaknesses, and drive meaningful, measurable improvements. You'll start by mapping and deeply understanding the existing pipeline, then move quickly to identify and execute near-term wins while simultaneously building the research and testing infrastructure needed to evaluate next-generation identity resolution approaches. A significant part of the role involves detecting, filtering, and eliminating bad data that degrades graph quality — requiring both investigative rigor and strong engineering capability. You'll also serve as Impact's technical counterpart with external identity and measurement partners — designing and running structured POCs, evaluating vendor solutions rigorously, and producing data-driven recommendations on build-vs-buy decisions that shape the long-term identity strategy.

Requirements

5+ years in data science, ML engineering, or applied research, with significant hands-on experience in identity graph construction, entity resolution, or large-scale identity matching in a production environment.
Deep, firsthand knowledge of how identity graphs are built, maintained, and evaluated — including deterministic and probabilistic linking, entity deduplication, graph schema design, and resolution at scale.
Demonstrated experience diagnosing and remediating data quality issues in large, complex datasets — identifying corrupted records, erroneous links, and systemic pipeline failures.
Proven ability to build and deploy production-grade pipelines independently; strong Python and SQL; solid software engineering fundamentals (testing, version control, observability).
Ability to design reproducible experiments, define meaningful evaluation metrics, and validate results before recommending adoption — in both internal and vendor evaluation contexts.
Ability to translate complex technical findings into clear strategic recommendations; experience presenting to cross-functional and leadership audiences.
Bachelor's in a quantitative field (CS, Statistics, Math, Engineering, or similar); Master's/PhD preferred.

Nice To Haves

Experience with graph databases and graph analytics (Neo4j, NetworkX, or similar) applied to identity or entity data.
Familiarity with ML-based entity resolution approaches: blocking strategies, embedding-based similarity, probabilistic matching, and hierarchical clustering.
Experience evaluating third-party identity resolution or data enrichment vendors; familiarity with the external identity and measurement partner landscape.
Exposure to affiliate marketing, ad tech, or performance marketing data environments and the identity challenges specific to those ecosystems.
Experience with real-time or near-real-time identity resolution and low-latency graph querying.
Familiarity with GCP tools (BigQuery, Vertex AI, Dataflow, Cloud Run) and/or Databricks/Spark for large-scale data processing.
Knowledge of privacy-preserving identity techniques (differential privacy, hashing, clean rooms) and their implications for graph construction.
Experience with device fingerprinting, cross-device identity, or cookie-less identity resolution approaches.

Responsibilities

Conduct a thorough, end-to-end mapping of Impact's current identity graph construction pipeline — data sources, entity linking logic, resolution rules, graph schema, and downstream consumers.
Document the architecture clearly and completely, creating a shared foundation of understanding for Data Science, Engineering, and Product stakeholders.
Identify structural weaknesses, coverage gaps, resolution failures, and data quality issues that degrade graph accuracy and completeness.
Produce a prioritized inventory of problems and opportunities, distinguishing quick wins from longer-horizon architectural improvements.
Identify and execute high-leverage, near-term improvements to the identity graph — targeting resolution accuracy, coverage, and data freshness without requiring full architectural overhaul.
Implement incremental enhancements to existing matching and linking logic; measure impact rigorously and communicate results to stakeholders.
Build the habit of continuous improvement into the graph pipeline: monitoring, alerting, and iterative refinement as ongoing practice rather than one-time effort.
Research, design, and implement methods to systematically identify, filter, and eliminate bad data in the identity graph — including corrupted identifiers, ghost entities, erroneous links, and stale or conflicting records.
Build detection pipelines that surface data quality issues proactively; define quality thresholds and SLOs for graph health.
Establish feedback loops that catch new bad data patterns as they emerge, preventing quality degradation over time.
Develop a robust testing and experimentation environment for evaluating cutting-edge identity resolution techniques — including probabilistic matching, deterministic linking, graph-based entity resolution, and ML-based approaches.
Design evaluation frameworks with clear, reproducible metrics for resolution precision, recall, coverage, and graph coherence.
Research and prototype emerging methods from the academic and industry literature; validate results rigorously before recommending adoption.
Maintain a living research agenda that tracks the state of the art in identity resolution and surfaces relevant advances to the team.
Engage directly with external identity and measurement partners to scope and execute structured POCs — evaluating vendor technologies against Impact's specific identity graph requirements.
Design rigorous, data-driven evaluation frameworks for vendor assessments; produce clear, evidence-based build-vs-buy recommendations for leadership.
Serve as Impact's technical counterpart in partner conversations — understanding vendor architectures deeply, asking sharp questions, and stress-testing vendor claims against real data.
Stay current on the external landscape for identity resolution, data enrichment, and measurement vendors; bring relevant developments to the team proactively.
Take improvements and new capabilities from research and POC to production, independently or in close partnership with Data Engineering — owning deployment, testing, monitoring, and iteration.
Write clean, well-tested, production-grade code; build pipelines that are maintainable and observable by the broader team.
Collaborate with MLOps and Platform Engineering to ensure production readiness: reliability, scalability, latency, and drift monitoring.
Translate complex identity graph findings — audits, quality analyses, vendor evaluations — into clear, actionable narratives for technical and non-technical audiences.
Present findings, tradeoffs, and recommendations in planning and leadership forums; communicate uncertainty and risk honestly.
Contribute to documentation and knowledge sharing that makes the identity graph understandable and trustworthy across the organization.

Benefits

Medical, Dental, and Vision insurance
Office-only catered lunch every Thursday, a healthy snack bar, and great coffee to keep you fueled
Flexible spending accounts
401(k)
Responsible PTO policy
Mental health and wellness benefit includes up to 12 fully covered therapy/coaching sessions per year, with additional dependent coverage.
Monthly gym reimbursement policy
Restricted Stock Units (RSUs)
Free Coursera subscription
PXA courses
26 weeks of fully paid leave for the primary caregiver
13 weeks fully paid leave for the secondary caregiver
Technology stipend to help you set up your home office
Monthly allowance to cover your internet expenses