About the role
Narvar is growing! We’re building the data infrastructure behind the post-purchase experiences of hundreds of millions of consumers. When a shopper tracks a package, initiates a return, or gets a delivery notification through one of our 1,500+ brand partners, our data systems are doing the work behind the scenes. We’re looking for an Analytics Engineer, Data to own internal analytics enablement — the bridge between our business teams and our data. You’ll translate questions from product, go-to-market, and executive partners into trusted metrics and build them into our semantic layer so they’re defined once and reused everywhere. At its core, this is a building role: you’ll expand our semantic layer and the Claude-powered agentic analytics on top of it — the metrics, models, and metadata that let our teams and Claude answer questions directly — so the organization can increasingly answer its own questions. It suits someone equally at home in a stakeholder conversation and in a data model, who measures success by how much self-serve they create. We work in an AI-native way: Claude is part of our daily analytics workflow, and much of this role is making our data ever more usable by it. What You’ll Work On Partner with product, go-to-market, and executive stakeholders — running discovery on ambiguous questions and scoping the metrics and data they actually need Raise data trust — adding the validation, definitions, and documentation that let users rely on the numbers and our tooling Expand and own our semantic / metrics layer — defining and maintaining metric definitions and models so analytics are consistent, trustworthy and reusable across the company Deliver self-serve and AI-accessible analytics — curated datasets, metrics, and reporting that internal partners and our agentic / LLM querying surface can answer on their own Ingest net new data — designing and building pipelines to bring in new sources such as GTM and product-usage data and modeling them for analytics What We’re Looking For We care about judgment and ownership over credentials. 3+ years in analytics engineering, data, or a closely related role, including ownership of metrics or data models that other teams rely on Deep SQL and hands-on data modeling — dimensional modeling, incremental transformations, and a feel for clean, maintainable models Proven experience building and expanding a semantic / metrics layer — its models, definitions, and context — that other teams adopt; you’ve owned what others depend on rather than consumed it Extensive hands-on experience using Claude/Codex for analytics — you’ve done substantive analytical work with it and know how to structure data, metrics, and metadata so it answers reliably The ability to stand up a new data source end to end — comfort with orchestration, APIs, and batch ETL, not just querying what already exists Excellent stakeholder communication — you can lead a conversation with a non-technical partner, walk away with a data spec, and explain a metric so people trust it A builder’s mindset — you’re motivated by creating durable, reusable metrics and self-serve infrastructure that scales beyond any single request Working knowledge of a cloud data warehouse (GCP / BigQuery preferred), a BI tool such as Looker, and Python for pipeline and tooling work Signals That You’ll Thrive Here These aren’t hard
