Senior Data Quality Analyst

Quality is core to Komodo Health’s mission to reduce the global burden of disease. As we scale the Healthcare Map® and AI-driven products, the need for independent, rigorous data quality oversight is critical.

We’re not hiring a traditional QA or testing specialist. We’re looking for an analytically rigorous, detail-oriented professional who knows pipelines can run flawlessly yet produce analytically wrong data—and that meeting specs doesn’t always mean meeting customer needs.

This is where the Senior Data Quality Analyst comes in. You’ll be the independent voice on data outputs—asking not just “did it run?” but “does this make sense for the customer?” Grounded in how customers use healthcare data, you’ll drive the analysis and quality monitoring needed to ensure it delivers real value.

The Senior Data Quality Analyst will own three standing operational responsibilities that are critical to Komodo’s weekly data delivery and major version release process:

• Data output validation: Run pre/post-release comparisons across key attributes to ensure changes meet data quality standards—not just engineering specs.
• Bug investigation support: Investigate issues from customer complaints and monitoring, document findings, and partner with engineering on root cause and resolution.
• Weekly publication review preparation: Assemble the execution summary, test coverage audit, and issue disposition list to support DPQ’s Monday release recommendation.

Komodo’s Data Product org is evolving—formalizing quality ownership, embedding automated QA, and establishing DPQ as the independent voice for every release. You’re joining at the moment that authority is being defined and put into practice.

The core of this role is independently assessing whether Komodo’s data outputs are analytically sound—ensuring the data tells the right story for customers, not just that the engineering works.

• Release validation: Design and run pre/post-release comparisons across key attributes (patient counts, claim volumes, fill rates, deduplication, payer attribution, provider coverage).
• Anomaly identification: Surface and document issues missed by automated tests—valid but suspicious patterns like demographic shifts, volume changes, or rule edge cases.
• Disposition recommendations: For each issue, assess what changed, customer impact, and recommend action (approve, conditional approve, hold, or escalate).
• Coverage documentation: Track what was tested, what passed, and accepted risks for each release—creating an auditable quality trail.

Data quality issues rarely surface clearly—this role requires the rigor to navigate ambiguity and the precision to communicate findings to both technical and non-technical stakeholders.

• Issue triage: Review and prioritize DPQ Jira issues, distinguishing data output problems (DPQ), pipeline failures (engineering), and cases needing joint investigation.
• Hands-on investigation:Query Snowflake to trace anomalies to source, validate against expectations, and rule out alternatives.
• Findings documentation: Produce clear reports outlining the issue, evidence, likely cause, and next steps for both technical and non-technical audiences.
• Resolution coordination: Partner with Data Engineering and Architects to drive resolution and verify fixes address the root issue.

Each week, Komodo publishes updated data. DPQ owns the release recommendation at the Monday meeting, and this role assembles the inputs that inform that decision.

• Pipeline execution summary: Compile a weekly record of which data pipelines ran, their completion status, any anomalies in run time or output volume, and a comparison against expected behavior.
• Test coverage audit: For each pipeline that ran during the week, document which quality checks were expected to execute and which did, surfacing any gaps in coverage that require manual review or escalation.
• Issue consolidation: Aggregate all quality issues raised during the week — from automated alerts, manual testing, and customer reports — into a single structured view with status, severity, and recommended disposition.
• Release recommendation package: Prepare the DPQ release recommendation document in advance of the Monday meeting, enabling DPQ leadership to review and present a confident, evidence-based recommendation.

This role rewards people who are energized by the gap between “it ran as designed” and “it’s actually right.” The ideal candidate is:

• Intellectually curious: Driven to investigate the unknown in complex data—never taking outputs at face value and always digging for root cause.
• Analytically rigorous: Go beyond error checks—spot missing data, unexpected trends, and subtle signals that something’s off.
• Customer-aware: Define quality by whether customers can answer real healthcare questions—not just whether tests pass.
• Precise communicator: Write clear, actionable reports for engineers and accessible summaries for non-technical partners.
• Operationally reliable: Consistently deliver a complete, on-time release package for the weekly publication meeting. 

• Experience: 6+ years of experience in data quality, data analysis, or analytics engineering — preferably in healthcare, life sciences, or another domain with complex, multi-source data.
• SQL proficiency: Strong SQL skills for large-scale analysis—joins, window functions, aggregations, and tracing data lineage. Snowflake preferred.
• Data investigation: Proven ability to work through ambiguous issues from signal to root cause—ruling out as well as ruling in.
• Structured QA process: Experience designing or executing pre/post release testing, including defining attributes, tolerances, and escalation criteria.
• Python (preferred): Comfortable using Python for analysis, validation, and light automation—able to read and adapt existing scripts.
• Healthcare data (required): Familiarity with claims data (medical, pharmacy, enrollment) and common quality patterns and failure modes.

• Ability to leverage AI tools (Gemini, Claude, Cursor, etc.) to enhance personal productivity, streamline workflows, content and visualization creation.

We want candidates to be clear-eyed about the scope of this role:

• This is not a data engineering role—you won’t build or maintain pipelines.
• This is not a software QA role—you won’t write unit tests, manage CI/CD, or review code.
• This is not a pure reporting role — you will be doing hands-on analytical work with real data, not summarizing dashboards others built.
• This is a data-first role where analytical judgment and quality intuition matter as much as execution.