Backend Engineer, Data

Description

The company builds AI agents for marketing leaders, working alongside marketers to automate workflows across analysis, campaign planning, activation, measurement and reporting.

The company has built its own stack from identity and enrichment through data integration to the agents that act on the data and report results. It has raised $20M.

A warehouse of consumer data only helps a marketer once they can sort, filter and chart it as quickly as they can ask the question. This role makes that possible inside the product.

The data team builds the consumer graph and each brand's first-party models, and data scientists train scores on top of them. You make all of this usable in the app: first-party fields that respond right away, scores on the pages marketers already use, segments that stay current, and an agent that turns a plain-English question into the right query on a brand's data.

This is an individual contributor or senior individual contributor role in product engineering, working daily with data engineering and data science. Your main counterpart on the warehouse side is the Staff Data Engineer, who owns the golden model and publishes the semantic views you build on.

In the first 90 days, new scores and first-party tables should reach the app through a repeatable system you have built. In the first 6 months, any first-party field should be sortable, filterable and chartable without a live warehouse query, and the in-app agent's query tool should run on your layer.

When product engineering has a data question, it should come to you rather than go unanswered.

Responsibilities

• Make first-party data fast in the app, so any connected field can be sorted, filtered and charted without a live warehouse query.
• Decide what gets precomputed, where it is stored and how it is kept fresh.
• Deliver propensity, LTV, churn and product-fit scores on person, segment and product pages, showing full distributions rather than a single number.
• Build dynamic segments that update as new data arrives, with stats ready for activation.
• Build the in-app agent's query tool on the semantic views the data platform team publishes, owning the path from a marketer's question to the matching rows.
• Own the product side of the sync, choosing what is precomputed and the route from Postgres to the synced client, weighing the cost and time of each hop.
• Extend the pipelines. Mostly you build on the data team's output, but when a feature needs a table that does not exist, you write the model or pipeline step yourself.
• Take features all the way through, which may mean a Python Lambda, a Zero schema change and a TanStack Start page in one piece of work.

Requirements

• Production experience with a cloud data warehouse. Snowflake experience is preferred.
• Production TypeScript, and backend services shipped on AWS such as Lambda.
• Keep up with AI tooling and with what changed in the latest releases of the packages you depend on.
• Side projects you built yourself that you can show.
• Based in New York City, or willing to relocate there.
• Likely fits include: a full-stack engineer at a data-heavy SaaS company who took on warehouse ownership; a product engineer at a warehouse-native analytics or activation company who has made Snowflake fast inside an app; a backend engineer who has shipped on a sync engine like Zero, Replicache or ElectricSQL and wants a larger dataset to test it, which is especially relevant; a data engineer who wants to own the feature their tables end up in; or someone who built something this year that changed how they think data should reach a user.
• No cover letter is needed. A demo of relevant work is required to apply.

Conditions

Base salary of $170,000 to $236,000, depending on level.

Work in person from the office at 300 Kent Avenue in Williamsburg, Brooklyn.

The company does not describe equal-opportunity terms in this posting.

Benefits

• Meaningful equity.
• Medical, dental and vision coverage.