Data Engineer, Measurement & Reporting

About Vibe At Vibe.co , we're reimagining how brands reach audiences in the age of streaming. We believe streaming TV is no longer just a brand awareness play, it's the next great performance marketing channel. We're building the infrastructure to unlock this $100B opportunity. Vibe.co provides an Audience First Streaming TV Advertising solution for marketers to unlock TV as a growth channel. Our all-in-one solution combines hyper-targeted audience segmentation, AI-powered insights and recommendations, real-time campaign optimization, and incrementality measurement, giving brands of all sizes the precision and transparency they've come to expect from social and search, but on TV. Trusted by over 10,000 brands, Vibe.co reaches more than 120 million households across 500+ apps and channels, delivering an average 250% return on ad spend and 20% sales lift. The company hit a $100 million revenue run rate in under two years, ranking among the ten fastest software companies to reach that milestone. Vibe.co 's investors include Hedosophia (an early backer of Spotify, Uber, and Airbnb), Elaia, Singular, QuantumLight (Revolut CEO Nik Storonsky's fund), and Illusian (Supercell CEO Ilkka Paananen's fund), as well as angel investors including Carolyn Everson, board member of The Walt Disney Company and Coca-Cola. Nirav Tolia, CEO of Nextdoor, sits on the company's board of directors. Founded in 2022, Vibe.co is widely recognized as the category-defining platform in streaming TV advertising, bringing the power of Meta and Google-style performance marketing to the fastest-growing segment in media. About the Role Vibe ingests more than 500,000 messages a second and stores several petabytes. Every advertiser who opens their dashboard expects their campaign numbers to be right, and current. You own the layer that makes that true. The platform side of the team builds the storage and compute. You run production on top of it: schema changes, data transformations, the reporting pipeline that feeds the client-facing UI. At this volume that is not maintenance work. A schema change is a distributed systems problem. A slow query is a customer-facing incident. We are already past what off-the-shelf systems handle, which is why this is a senior seat and not a junior one. You get production ownership from week one, direct access to the engineers who built the infrastructure underneath you, and problems that don't have a Stack Overflow answer. What You'll Do Ship schema changes across petabyte-scale tables without breaking downstream consumers or advertiser reporting Make the reporting stack faster and cheaper. ClickHouse and Cube serve the client UI, so query latency and compute spend are yours to defend Turn new product features into production data models, from raw Kafka and Iceberg through to what the advertiser sees Trace discrepancies back through the stack to the source system, then fix the cause instead of the symptom Catch scaling and data-quality problems before they page someone at 2am Push back when a request costs more than it's worth. "Not yet" and "not this way" are answers we expect to hear from you What You Need 5+ years in production data engineering. You have owned pipelines real users depended on, with on-call attached Strong SQL and schema design at a volume where the design choice actually mattered Production experience with an orchestrator (Dagster, Airflow, Prefect) plus dbt or an equivalent transformation framework One optimization win you can walk us through end to end: what was slow or expensive, what you changed, what the number went from and to, and why it worked Systems-level depth. Our technical round is the team drilling into how things work underneath, not a puzzle screen The judgment to turn a vague request into a concrete plan, and the comfort to do it fast Nice to Haves ClickHouse in production, or another column store at serious scale (Druid, Pinot, BigQuery) with real w