About Preql

Preql helps enterprises clean, unify, and govern messy internal data so it actually works for AI, analytics, and reporting. We work with large organizations navigating complex data environments and high-stakes operational workflows. Based in New York, our team comes from data infrastructure, AI, and enterprise software.

How we work

We’re a small team with little bureaucracy. Leadership expects individuals to take ownership, move quickly, and make good decisions for the company with support from their teammates. The curious do well here, are comfortable operating in ambiguity, and are willing to form opinions and act on their convictions instead of waiting for instructions.

You will sit inside customer environments, learn how a specific finance organization actually closes its books and plans its year, and build the semantic models that make that work. You will be the person who understands both a customer's GL and our platform internals well enough to get the numbers right and defend them to a controller.

This is not a support role and it is not pure services. Every deployment you run should make the next one faster. The work you do by hand in month one should be a product capability by month six. You will be the loop between what customers need and what we build.

What you will own

• The business outcomes for a portfolio of enterprise accounts, from kickoff through production and expansion

• Semantic models for finance logic: revenue recognition, cost allocation, GL and cost center hierarchies, headcount and driver based planning

• Source integration and mapping across ERPs, planning systems, and warehouses, including the reconciliation problems that surface once real data lands

• Working sessions with controllers, FP&A leads, and customer data teams, translating between finance language and data models

• The judgment call on what is a modeling problem, a source data problem, or a product gap, and routing each one to the right place

• A steady stream of product feedback backed by specifics, not anecdotes, so engineering builds against real customer friction

• Reusable models, templates, and documentation that shrink time to value on every subsequent account

What success looks like

• 90 days: you have taken an account from install to first trusted output, and you can explain any number in a customer's reporting back to its source

• 6 months: time to first value for a comparable account has dropped measurably because of models and assets you built, and customers ask for you by name

• 12 months: the delivery playbook is yours, expansion conversations start with work you did, and the next engineers we hire ramp against what you wrote

What we are looking for

• 5+ years building with data in production, with deep SQL fluency and comfort in Python

• Direct experience with cloud warehouses (Snowflake, Databricks, BigQuery) and transformation tooling (dbt or equivalent)

• Real working knowledge of financial data. You know why the finance team's definition of revenue is different from the data team's, and you have modeled a chart of accounts, an allocation, or a close process before

• Experience working directly with enterprise customers, including the parts that are uncomfortable: scoping, pushing back, and delivering bad news early

• High tolerance for ambiguity. Early accounts will not have a playbook, and you will write the playbook

• Judgment about when to solve something for one customer and when to solve it for all of them

Strong signals

• Familiarity with ERP and planning systems (NetSuite, Workday, SAP, Oracle)

• Background in consulting, solutions architecture, or professional services at a data or AI company

• You have worked with regulated buyers

• You have been the first or second technical hire on a customer facing team

• Time spent inside a finance or accounting function, or close enough to one to have felt a close