About the Role :

Applied Retention is building an AI decisioning layer for retention marketing. You are the founding engineer for its data substrate and nightly runtime, working directly with the CTO. You will design the schema, build the ingestion, write the nightly computations the product decides on, and make it correct enough that brands bet real marketing spend on it. The role is roughly 70% data engineering and 30% backend.

What you'll build :

- Ingestion pipelines, multi-tenant from multiple async data sources

- Nightly batch computation over millions of events as set-based SQL and dbt models

- A multi-tenant data model where one brand's data never affects another's run

- Backfill, testing infrastructure and replays as designed capabilities

- Vendor integrations built for failure : retries, idempotency, reconciliation

Requirements :

- 4 - 6 years building and operating production systems

- Deep PostgreSQL : schema design, query plans, batch performance on tables in the tens of millions of rows

- Owned a data pipeline in production end to end : ingestion, transformation, reconciliation, serving

- Set-based thinking : multi-step SQL, window functions, percentiles, incremental computation; you've turned raw event or order data into cohort, retention, or funnel metrics

- Production Python, any framework

- You test batch jobs and reconciliations

- Clear written communication

- Able to start within 30 days

Nice to have :

- dbt

- Django in production

- AWS Stack : Python, Django, PostgreSQL, dbt, React with TypeScript, AWS.