Metadata Specialist Full Time NEW

About the Team

Our New Verticals Strategy & Operations team is re-inventing the way consumers think about grocery, convenience, alcohol and retail delivery through both our first-party and third-party operations businesses. The Global Catalog team plays a critical role in shaping DoorDash’s e-commerce strategy, leading some of the company’s most strategic programs to drive catalog excellence. Their work is highly visible and impactful-spanning the entire New Verticals catalog across DoorDash, Deliveroo, and Wolt-and influences tens of millions of items on the platform.

About the Role

We are looking to hire an experienced Taxonomist to join the Global New Verticals Catalog team across DoorDash, Wolt, and Deliveroo. This role will focus on developing and improving the processes and models that power DoorDash’s three-sided marketplace-Merchants, Dashers, and Customers-as well as its emerging fourth side, Advertisers, with the goal of empowering and growing local economies. Achieving this requires a deep understanding of local businesses, the products and services they offer, and the consumers who interact with them. A strong grasp of the global ecosystem is essential to building and mapping complex taxonomies and entity relationships within local economies. This includes the design, implementation, and governance of large-scale taxonomies to accelerate product development and deliver high-quality consumer experiences.

As a Taxonomist on the Global Catalog team, this role will be responsible for Product Categories and Brands taxonomies, as well as the broader attribute platform. The position will work closely with Strategy & Operations, Machine Learning, Engineering, Product, and Data Science partners.

You’re excited about this opportunity because you will

• Build and maintain the metadata schema, attribute systems, and entity relationships within a global ontology platform

• Design and maintain taxonomies across consumer domains including grocery, alcoholic beverages, apparel and more, ensuring alignment with attribute models

• Develop and apply best practices for taxonomy and ontology modeling, including hierarchy design, attribute standardization, and relationship definition .

• Identify and address gaps in catalog quality, completeness, and discoverability across international markets.

• Partner cross-functionally with Product, Operations, Analytics, and regional teams to design and operationalize taxonomy and ontology solutions globally

• Evolve and scale taxonomy and ontology systems for New Verticals, aligning and unifying structures across DoorDash, Wolt and Deliveroo markets

• Contribute to strategic initiatives that strengthen merchant catalogs and improve the customer shopping experience across international markets.

We’re excited about you because

• You solve problems from first principles. You’re excited to think innovatively, and you resist “this is how it’s always been done” thinking.

• You are an operator. You don’t hesitate to get in the weeds and operate at the lowest level of detail, but you’re also able to think long-term. You will experience our operation first hand, to inform your strategy.

• You’re analytical. You can work with data across multiple sources, ask the right questions of a dataset, and communicate findings clearly, using whatever tools get you there fastest (e.g., SQL, BI tools or AI-assisted analysis).

• You have 5+ years of experience in developing complex taxonomies, ontologies, or structured classification systems in real-world applications.

• You have experience with taxonomy or ontology management tools (Graphite, PoolParty, Graphologi, etc.) and have designed taxonomies to support navigation, search, discovery, and analytics at scale.

• You have completed a Masters Degree in either Library and Information Science, Linguistics, Philosophy, or a related field.

Nice to Haves

• Experience with AI tools such as Claude Code and Cursor to accelerate problem-solving, analysis, or workflow execution.

• Proficiency in SQL.

• Familiarity with graph-based or triple-store systems (e.g., RDF, SPARQL) is a plus