At Opendoor, pricing is at the core of our product — our models directly influence high-stakes decisions around real estate transactions across the country. We are looking for a mid-level Software Engineer to join our Pricing & ML team. This is a unique opportunity to work side-by-side with applied researchers and modelers to build and productionize the machine learning models that power our pricing engine.
Requirements
- Has 2–4 years of experience in software engineering, ideally with exposure to machine learning workflows.
- Is comfortable navigating data pipelines, model training pipelines, and production environments.
- Is fluent in writing maintainable, modular, and testable Python code.
- Experience working on ML systems in business-critical environments (e.g., pricing, forecasting, logistics).
- Familiarity with tools like MLflow, Airflow, Delta Lake, or Spark.
- Experience monitoring model performance in production (e.g., drift detection, quality alerts).
Responsibilities
- Work closely with researchers and analysts to assist the development of pricing models, and convert prototypes into robust, production-grade code.
- Own and operate model pipelines end-to-end — including training, validation, versioning, deployment, and monitoring.
- Contribute to shared ML infrastructure and tooling as needed, while staying focused on solving business-critical problems.
- Support day-to-day pricing model operations and address real-world ML lifecycle challenges like retraining, data drift, and model decay.
- Navigate real estate-specific ML challenges (heterogeneous data, complex seasonality, sparse regions, high financial stakes).
- Drive engineering best practices within the ML codebase — maintainable, testable, and scalable systems.
Other
- Enjoys working closely with data scientists and researchers and has strong collaboration and communication skills.
- Is motivated by impact and learning — not just building infrastructure for others.
- An interest in real estate or other messy, high-stakes domains with imperfect data.
