Remote Senior Applied Machine Learning Engineer - Applied Machine Learning Team

Rocket Mortgage•Seattle, WA

1d•$164,300 - $258,100•Remote

About The Position

The Applied Machine Learning group at Redfin works towards redefining real estate in the customer’s favor using machine learning. We work on foundational problems in the real estate space including recommendations (“Where should I live?”) and price estimation (“How much is a home worth?”). The Brokerage Recommendations product alone, which we power drives 27% of all traffic to Redfin platforms. We have real estate data at a national level and work across various domains in machine learning using large-scale multi-modal property data (documents, images, text, video, 3D scans etc.). Our team also owns and maintains end-to-end production-grade large-scale machine learning infrastructure and systems serving hundreds of millions of consumers. As a Senior Machine Learning Engineer for the Applied Machine Learning Team, you’ll breathe life into our research by transforming prototypes into high-performance production systems, building the automated MLOps pipelines and real-time optimizations that keep our models running at scale. As the strategic bridge between research and engineering, you’ll own the health of our valuation and recommender systems to ensure they remain fast, reliable, and impactful for millions of users.

Requirements

5+ years of software engineering experience, with at least 2 years specifically focused on deploying and scaling machine learning models in production environments.
Highly proficient in Python and capable of writing production-grade, modular code.
Deep understanding of the end-to-end ML lifecycle, including training versus inference workflows, feature stores, and model versioning.
Competent with ML frameworks such as PyTorch, TensorFlow, or Scikit-learn.
Ability to effectively debug, tune, and optimize models for inference.
Competent with monitoring, observability, and production maintenance.
Ability to effectively set up and manage logging, metrics, and alerting pipelines, debug production issues, and ensure system reliability and performance in high-scale feature store.
Hands-on experience with Docker and Kubernetes.
Understanding of how to deploy, manage, and scale containers within a cluster environment.
Experience implementing model monitoring and operational rigor, including tracking data drift and latency, as well as utilizing A/B testing frameworks to validate model performance in the wild.
Proficient in SQL and familiar with distributed data processing tools like Spark or Kafka to ensure that data reaching the model is high-quality and consistent with training distributions.

Responsibilities

Productionize models by converting research-grade code into performant, clean, and maintainable production systems.
Implement MLOps best practices, including CI/CD for machine learning, automated retraining pipelines, and robust model versioning.
Optimize models for inference to ensure high-speed performance and efficiency in real-time environments.
Monitor models in production, proactively identifying and mitigating issues related to data drift, concept drift, and system latency.
Co-create the next generation of data-driven insights for automated valuation models (AVM) and recommender systems.
Identify and implement iterative improvements to the machine learning models that power production-scale, customer-facing experiences.
Serve as a technical bridge, assisting other engineers and stakeholders in understanding and applying data science methodologies and findings across the organization.
Build data products and analytical tools that drive critical business metrics and revenue growth, directly impacting the home buying and selling experience.